Data Methods for Health Research Assignment

Loading of standard modules

import pandas as pd
import numpy as np

%matplotlib inline
import random
import matplotlib.pyplot as plt
import matplotlib.colors as colors
import matplotlib.ticker as mtick

from math import floor
from pandasql import PandaSQL

import numpy.polynomial.polynomial as poly

Read in Data

Practice Address Data

############################################
# Read in the GP address data as a CSV file 
# HSCIC GP Practice Prescribing (Dec/2015)
############################################

# Store the url of the data as a variable
url_GP_address= 'https://s3.eu-west-2.amazonaws.com/dmhr-data/practices_Dec2015.csv'

#Introduce column headings as original data has none
col_heads_GP_address = ['PERIOD', 'PRACTICE', 'PRACTICE NAME', 'ADDRESS_1', 
                        'ADDRESS_2', 'ADDRESS_3', 'ADDRESS_4', 'POSTCODE']

#Set up a DataFrame using a subset of the original data.
df_GP_address_ORIG = pd.read_csv(url_GP_address, names=col_heads_GP_address, index_col=False, 
                            usecols=['PRACTICE', 'PRACTICE NAME','POSTCODE'])

#Remove trailing spaces from data.
df_GP_address_ORIG = df_GP_address_ORIG.applymap(lambda x: str(x).strip())

df_GP_address_ORIG.head(1)

Read in Data

Medications Prescribed with price & spend Data

###############################################
# Read in the GP prescribing data as a CSV file
# HSCIC GP Practice Prescribing (Dec/2015)
###############################################

# Store the url of the data as a variable
url_GP_prescribing= 'https://s3.eu-west-2.amazonaws.com/dmhr-data/prescribing_Dec2015.csv'

# Create column headings for DataFrame
col_heads_GP_prescribing = ['SHA', 'PCT', 'PRACTICE', 'BNF CODE', 'BNF NAME', 'ITEMS', 
                            'NIC (£)', 'ACT COST (£)', 'QUANTITY', 'PERIOD','BLANK']

#Set up a DataFrame
df_GP_prescribing_ORIG = pd.read_csv(url_GP_prescribing,header=0,
                                          names=col_heads_GP_prescribing, index_col=False)

df_GP_prescribing_ORIG.head(1)

# Note: Before clean, GP data column headings had too many spaces and one empty column

Read in Data

Number of Patients Data

###############################################################################
# Read in the number of registered patients in GP practices as a CSV data file 
# HSCIC GP demographic information
###############################################################################

# Store the url of the data as a variable
url_GP_patient_nbrs = 'https://digital.nhs.uk/media/28273/Numbers-of-Patients-Registered-at-a-GP-Practice-Jan-2016-GP-Practice-and-quinary-age-groups/Any/gp-reg-patients-prac-quin-age'


#Set up a DataFrame using a subset of the original data.
df_GP_patient_nbrs_ORIG = pd.read_csv(url_GP_patient_nbrs, index_col=False, 
                                 usecols=['GP_PRACTICE_CODE', 'Total_All'])

#Rename column headings
df_GP_patient_nbrs_ORIG.rename(columns={'GP_PRACTICE_CODE':'PRACTICE','Total_All':'TOTAL NUMBER OF PATIENTS'}, inplace=True)


#Remove trailing spaces from data.
df_GP_patient_nbrs_ORIG['PRACTICE'] = df_GP_patient_nbrs_ORIG['PRACTICE'].str.strip()

df_GP_patient_nbrs_ORIG.head(1)

Assignment A

A (0) Prelimaries

#Bolton postcodes
postcodes_list_bolton=['BL1','BL2','BL3','BL4','BL5','BL6','BL7']

#DataFrame for Bolton Postcodes
df_postcodes_bolton = pd.DataFrame(np.array(postcodes_list_bolton).reshape(7,1), 
                                   columns = ['BOLTON POSTECODES'])

#Cambridge postcodes
postcodes_list_cambridge=['CB1','CB2','CB3','CB4','CB5','CB21','CB22','CB23','CB24','CB25']

#DataFrame for Cambridge Postcodes
df_postcodes_cambridge = pd.DataFrame(np.array(postcodes_list_cambridge).reshape(10,1), 
                                      columns = ['CAMBRIDGE POSTECODES'])

# Combine 'df_GP_prescribing_ORIG' with 'df_GP_address_ORIG'

df_GP_prescribing_merge_with_address = pd.merge(df_GP_prescribing_ORIG, df_GP_address_ORIG, how='left',
                                          left_on='PRACTICE', right_on='PRACTICE')

# Extract from "df_GP_prescribing_merge_with_address" the first 4 characters from 'POSTCODE',
# rename column heading as 'SHORT POSTCODE' & remove blanks

df_postcode_extract= pd.DataFrame(df_GP_prescribing_merge_with_address['POSTCODE'].astype(str).str[:4]).rename(columns = {'POSTCODE':'SHORT POSTCODE'})
df_postcode_short=df_postcode_extract['SHORT POSTCODE'].str.replace(' ', '')


df_GP_practice_prescribing = pd.concat([df_GP_prescribing_merge_with_address, df_postcode_short], axis=1)

#df_GP_practice_prescribing.head(1)

#######################################
# Create Bolton & Cambridge DataFrames
######################################

# Bolton prescribing data only 
# dropping columns 'BOLTON POSTECODES' &'SHORT POSTCODE' using code: df.drop(['B', 'C'], axis=1)

df_GP_practice_prescribing_bolton = pd.merge(df_postcodes_bolton, df_GP_practice_prescribing, how='left',
                                             left_on='BOLTON POSTECODES', right_on='SHORT POSTCODE').drop(['BOLTON POSTECODES', 'SHORT POSTCODE'], axis=1)

# Cambridge prescribing data only 
# dropping columns 'CAMBRIDGE POSTECODES' &'SHORT POSTCODE' using code: df.drop(['B', 'C'], axis=1)

df_GP_practice_prescribing_cambridge = pd.merge(df_postcodes_cambridge, df_GP_practice_prescribing, how='left',
                                             left_on='CAMBRIDGE POSTECODES', right_on='SHORT POSTCODE').drop(['CAMBRIDGE POSTECODES', 'SHORT POSTCODE'], axis=1)
#df_GP_practice_prescribing_bolton.head(1)

Assignment A

A (1) i. Identification of all GP practices

A (1) ii. Production of a table with the "total number of prescriptions" and their "total actual cost"

#This is for the total number of prescriptions Summing over "ITEMS" column
# & the total cost of prescriptions Summing over "ACT COST" column

df_Rx_nbr_spend=pd.DataFrame(df_GP_practice_prescribing_bolton.groupby('PRACTICE')['ITEMS','ACT COST (£)'].sum())

df_Rx_nbr_spend.columns=['TOTAL NUMBER OF PRESCRIPTIONS','TOTAL COST OF PRESCRIPTIONS (£)'] # Name the column

df_Rx_nbr_spend.reset_index(inplace=True) # Reseting the index


#A DataFrame with only "PRACTICE" & "PRACTICE NAME" from 'df_GP_practice_prescribing_bolton'

df_practice_name_code=df_GP_practice_prescribing_bolton.loc[:,'PRACTICE':'PRACTICE NAME':8]

#Merge the two dataframes 'df_practice_name_code' and 'df_Rx_nbr_spend'

df_QA_1_merge=pd.merge(df_practice_name_code,df_Rx_nbr_spend).drop_duplicates(subset=None, keep='first', inplace=False)


df_QA_1_sorted=df_QA_1_merge.sort_values(['PRACTICE'], ascending=[True]) #Sort by PRACTICE code
df_QA_1=df_QA_1_sorted.reset_index(drop=True) # Reseting the index


#######################
# Results table styled
#######################

df_QA_1_styled=df_QA_1.style.format({
    'TOTAL NUMBER OF PRESCRIPTIONS': '{:,.0f}'.format,
    'TOTAL COST OF PRESCRIPTIONS (£)': '{:,.0f}'.format})

df_QA_1_styled

Assignment A

A (2) i. The top ten most frequently prescribed medications across all Bolton practices and their total actual costs

A (2) ii. The top ten least frequently prescribed medications across all Bolton practices and their total actual costs

##################################################################################
# Create a truncated version of the DataFrame "df_GP_practice_prescribing_bolton"
# in which columns with numerical values are summed
# giving totals for "ITEMS" "NIC" "ACT COST" "QUANTITY" for each medication
##################################################################################

#Extract drugs and corresponding numerical values

temp_Rx=df_GP_practice_prescribing_bolton.loc[:,'BNF NAME':'QUANTITY':]

#sum numerical values, give column headings, insert index values

temp_Rx_summary=pd.DataFrame(temp_Rx.groupby('BNF NAME').sum())
temp_Rx_summary.columns=['TOTAL ITEMS DISPENSED','TOTAL NET INGREDIENT COST (£)','TOTAL ACTUAL COST (£)','TOTAL QUANTITY DISPENSED']
temp_Rx_summary.reset_index(inplace=True)


######################################################################################################
# A DataFrame with only "BNF CODE" & "BNF NAME" from the DataFrame "df_GP_practice_prescribing_bolton"
######################################################################################################

temp_Rx_code=df_GP_practice_prescribing_bolton.loc[:,'BNF CODE':'BNF NAME':]

#####################################################################
# combine "temp_Rx_code" and "temp_Rx_summary" and remove duplicates
#####################################################################

combine_temp_Rx_code_temp_med_summary=pd.merge(temp_Rx_code,temp_Rx_summary).drop_duplicates(subset=None, keep='first', inplace=False)

#reindex 

Rx_totals_bolton_full=combine_temp_Rx_code_temp_med_summary.reset_index(drop=True)

#drop columns not asked for 
temp_Rx = Rx_totals_bolton_full.loc[:, ['BNF CODE','BNF NAME','TOTAL ITEMS DISPENSED','TOTAL ACTUAL COST (£)']]

##############################################################################
# Getting top ten most frequently prescribed medications across all practices
# and the total actual cost
##############################################################################

#Identify top 10 most dispensed

top_Rx_bolton_all=temp_Rx.sort_values(['TOTAL ITEMS DISPENSED'], ascending=False)

top_10_Rx_bolton=top_Rx_bolton_all.head(10)

top_10_Rx_bolton.index=['Most prescribed', '2nd most prescribed','3rd', 
                        '4th','5th','6th','7th','8th', '9th','10th']

#######################
# Results table styled
#######################

top_10_Rx_bolton_styled=top_10_Rx_bolton.style.format({
    'TOTAL ITEMS DISPENSED': '{:,.0f}'.format,
    'TOTAL ACTUAL COST (£)': '{:,.0f}'.format})

top_10_Rx_bolton_styled

##########################################################################
# Barchart comparing TOTAL ITEMS DISPENSED with TOTAL ACTUAL COST (£) for 
# top 10 most frequently prescribed medications
##########################################################################

#Create tuples from the dataframe "top_10_Rx_bolton" for barchart

df_for_tuples_a2i=top_10_Rx_bolton.reset_index(drop=True)

med_name_a2i=tuple(df_for_tuples_a2i['BNF NAME'].tolist())
med_nbr_item_a2i=tuple(df_for_tuples_a2i['TOTAL ITEMS DISPENSED'].tolist())
med_actual_cost_a2i=tuple(df_for_tuples_a2i['TOTAL ACTUAL COST (£)'].tolist())

ind = np.arange(len(med_nbr_item_a2i))  # x locations for the groups
width = 0.4  # width of the bars

fig, ax = plt.subplots(figsize=(9, 6))

rects1 = ax.bar(ind - width/2,med_nbr_item_a2i, width,color='olive', label='Total number of prescriptions')
rects2 = ax.bar(ind + width/2, med_actual_cost_a2i, width, color='orange', label='Total actual cost (£)')

plt.xticks(rotation=85)
ax.set_title('Barchart comparing the number of items dispensed vs total actual cost of the 10 most frequently prescribed medications', fontsize=14, fontweight='bold')
ax.set_xticks(ind)
ax.set_xticklabels(med_name_a2i, fontsize=11)
ax.legend()

##########################
# Format X & Y tick values
##########################

fmt = '{x:,.0f}'

tick = mtick.StrMethodFormatter(fmt)
ax.yaxis.set_major_formatter(tick) 

plt.show()

###########################################################################
# Getting ten least frequently prescribed medications across all practices
# and the total actual cost
###########################################################################

bottom_Rx_bolton=temp_Rx.sort_values(['TOTAL ITEMS DISPENSED', 'TOTAL ACTUAL COST (£)'], 
                                        ascending=[True, True])

bottom_10_Rx_bolton=bottom_Rx_bolton.head(10)

bottom_10_Rx_bolton.index=['Least prescribed','2nd least prescribed', '3rd', 
                           '4th','5th', '6th','7th','8th', '9th','10th']

#######################
# Results table styled
#######################

bottom_10_Rx_bolton_styled=bottom_10_Rx_bolton.style.format({
    'TOTAL ITEMS DISPENSED': '{:,.0f}'.format,
    'TOTAL ACTUAL COST (£)': '{:,.2f}'.format})

bottom_10_Rx_bolton_styled
#Note Also sorted on 'TOTAL ACTUAL COST (£)' as there were >1500 meds where only 1 item was prescribed

#Sum the total amount spent on the top 10 prescribed meducations in Bolton

df_Rx_most_least = pd.DataFrame([top_10_Rx_bolton['TOTAL ACTUAL COST (£)'].sum(),
                  bottom_10_Rx_bolton['TOTAL ACTUAL COST (£)'].sum()], 
                  columns=['Total Amount Spent across all Bolton (£)'], 
                  index=['10 most frequently prescribed medications',
                         '10 least frequently prescribed medications'])

#######################
# Results table styled
#######################

df_Rx_most_least_styled=df_Rx_most_least.style.format({'Total Amount Spent across all Bolton (£)': '{:,.0f}'.format})

df_Rx_most_least_styled

Assignment A

A (2) iii. Comparing the most/least frequently prescribed medications in Bolton practices to the overall actual costs of each practice and to the overall actual costs of the entire city

#######################################################################################
# Comparing actual costs on most frequently prescribed medications across all practices
# with the overall actual costs of each practice
#######################################################################################

transpose_top_bolton=pd.DataFrame(top_10_Rx_bolton.transpose())

transpose_top_bolton_Shrt= transpose_top_bolton.drop(['BNF CODE','BNF NAME','TOTAL ITEMS DISPENSED'])

############################################
#Expand so that as big as QA_1
############################################

no_rows_QA_1=len(df_QA_1.index)

temp_long1=pd.concat([transpose_top_bolton_Shrt]*no_rows_QA_1)

temp_long=temp_long1.reset_index()


shrt_QA_1=df_QA_1.filter(['PRACTICE','TOTAL COST OF PRESCRIPTIONS (£)'], axis=1) #This drops columns


##########################################
# need to combine shrt_QA_1  temp_long
##########################################

result_temp=shrt_QA_1.join(temp_long)

compare_most=result_temp.drop(['index'], axis=1)

#####################
#finding proportions
#####################
print("""

The table below shows the amounts spent on each of the 10 most frequently prescribed medications 
in Bolton as a percentage of the overall monthly spend by each of the GP practices in Bolton.

""")

compare_most.loc[:,"Most prescribed":"10th"] = compare_most.loc[:,"Most prescribed":"10th"].div(compare_most["TOTAL COST OF PRESCRIPTIONS (£)"], axis=0)
compare_most = compare_most.rename(columns={'TOTAL COST OF PRESCRIPTIONS (£)': 'Total Monthly Spend on all Prescriptions (£)'})
#compare_most.head()

#######################
# Results table styled
#######################

compare_most_styled=compare_most.style.format({'Total Monthly Spend on all Prescriptions (£)': '{:,.0f}'.format,
'Most prescribed': '{:,.1%}'.format,'2nd most prescribed': '{:,.1%}'.format,
'3rd': '{:,.1%}'.format,'4th': '{:,.1%}'.format,'5th': '{:,.1%}'.format,'6th': '{:,.1%}'.format,
'7th': '{:,.1%}'.format,'8th': '{:,.1%}'.format,'9th': '{:,.1%}'.format,'10th': '{:,.1%}'.format})

compare_most_styled


The table below shows the amounts spent on each of the 10 most frequently prescribed medications 
in Bolton as a percentage of the overall monthly spend by each of the GP practices in Bolton.

## # Comparing actual costs on most frequently prescribed medications across all practices
# with the overall actual costs of the city of Bolton as a whole
#######################################################################################

# Total Spend on all Prescriptions for the Month
df_bolton_sum_total = pd.DataFrame([{'All medications prescribed in Bolton':df_QA_1['TOTAL COST OF PRESCRIPTIONS (£)'].sum()}])

#total spend on top 10 prescriptions
df_Rx_most=df_Rx_most_least[0:1].reset_index(drop=True).rename(columns={'Total Amount Spent across all Bolton (£)': 'Top 10 most frequently prescribed medications'}) 

#Use temp column fix to merge two tables with no common columns
df_bolton_sum_total ['tmp'] = 1
df_Rx_most['tmp'] = 1

df_Rx_bolton_and_most= pd.merge(df_bolton_sum_total, df_Rx_most, on=['tmp'])

df_Rx_bolton_and_most=df_Rx_bolton_and_most.drop('tmp', axis=1) # drop the temp column

#Use temp column fix to merge two tables with no common columns
df_Rx_bolton_and_most['tmp'] = 1
transpose_top_bolton_Shrt['tmp'] = 1

Compare_city_with_top_Rx_temp = pd.merge(df_Rx_bolton_and_most, transpose_top_bolton_Shrt, on=['tmp'])


Compare_city_with_top_Rx_temp=Compare_city_with_top_Rx_temp.drop('tmp', axis=1) # drop the temp column

Compare_city_with_top_Rx_temp2=pd.concat([Compare_city_with_top_Rx_temp]*2)

Compare_city_with_top_Rx_temp3=Compare_city_with_top_Rx_temp2.reset_index()

############################################################
#Below divides the new row by total bolton giving proportions
#############################################################

Compare_city_with_top_Rx_temp3.loc[1] = Compare_city_with_top_Rx_temp3.loc[1] / df_QA_1['TOTAL COST OF PRESCRIPTIONS (£)'].sum()

#Below renames indices

Compare_city_with_top_Rx_temp3.index=['Amount Spent for the Month (£)',
                                      'as a percentage of Total Bolton Prescription Spend']

#need to drop index column 

Compare_city_with_top_Rx_temp=Compare_city_with_top_Rx_temp3.drop('index', axis=1)

Compare_city_with_top_Rx=Compare_city_with_top_Rx_temp.T

#######################
# Results table styled
#######################

Compare_city_with_top_Rx_styled=Compare_city_with_top_Rx.style.format(
    {'Amount Spent for the Month (£)': '{:,.0f}'.format,
'as a percentage of Total Bolton Prescription Spend': '{:,.2%}'.format,})

Compare_city_with_top_Rx_styled

Assignment A

A (3) i. The top ten most expensive medications prescribed in Bolton

##############################################################
## Finding cost of Rx per item. Found using (ACT COST / ITEMS)
##############################################################

temp_Rx=df_GP_practice_prescribing_bolton.loc[:,'BNF CODE':'ACT COST (£)':]
temp_Rx.loc[:,"NIC (£)":"ACT COST (£)"] = temp_Rx.loc[:,"NIC (£)":"ACT COST (£)"].div(temp_Rx["ITEMS"],axis=0)

temp_Rx = temp_Rx.rename(columns={'ACT COST (£)': 'ACT COST PER ITEM (£)'})

Rx_cost_per_item_bolton=temp_Rx.drop(['ITEMS', 'NIC (£)'], axis=1)

#######################################################
## Sort "Rx_cost_per_item_bolton" highest values first
#######################################################

Rx_cost_per_item_bolton_sorted=Rx_cost_per_item_bolton.sort_values(['ACT COST PER ITEM (£)'], 
                                                                    ascending=[False])

#################################################
## DataFrame with only top 10 most expensive Rx
################################################


top_10_Rx_cost_per_item_bolton=Rx_cost_per_item_bolton_sorted.head(10)
top_10_Rx_cost_per_item_bolton.index=['Most expensive', '2nd most expensive','3rd', 
                                      '4th','5th', '6th', '7th','8th', '9th','10th']

#######################
# Results table styled
#######################

top_10_Rx_cost_per_item_bolton_styled=top_10_Rx_cost_per_item_bolton.style.format(
    {'ACT COST PER ITEM (£)': '{:,.0f}'.format})

top_10_Rx_cost_per_item_bolton_styled

##################################################################################
# Barchart showing ACT COST PER ITEM (£) for the top 10 most expensive medications
##################################################################################

#Create tuples from the dataframe "top_10_Rx_cost_per_item_with_total_spend_bolton" for barchart

df_for_tuples_a3i=top_10_Rx_cost_per_item_bolton.reset_index(drop=True)

med_name_a3i=tuple(df_for_tuples_a3i['BNF NAME'].tolist())
med_cost_per_item_a3i=tuple(df_for_tuples_a3i['ACT COST PER ITEM (£)'].tolist())


ind = np.arange(len(med_cost_per_item_a3i))  # x locations for the groups
width = 0.4  # width of the bars

fig, ax = plt.subplots(figsize=(9, 6))

rects1 = ax.bar(ind,med_cost_per_item_a3i, width,color='c', label='Cost per item')


plt.xticks(rotation=85)
ax.set_title('Barchart showing the cost per item for the 10 most expensive medications prescribed in Bolton', fontsize=14, fontweight='bold')
ax.set_xticks(ind)
ax.set_xticklabels(med_name_a3i, fontsize=11)


##########################
# Format X & Y tick values
##########################

fmt = '£{x:,.0f}'

tick = mtick.StrMethodFormatter(fmt)
ax.yaxis.set_major_formatter(tick) 

plt.show()

Assignment A

A (3) ii. The total actual cost of the top ten most expensive medications prescribed in Bolton

#################################################
## Join DataFrame with top 10 most expensive and 
## a shortened version of numerical summary table
#################################################

#short version of numerical summary table calling it "temp_join"

temp_join=Rx_totals_bolton_full.drop(['BNF NAME', 'TOTAL ITEMS DISPENSED',
                                      'TOTAL NET INGREDIENT COST (£)','TOTAL QUANTITY DISPENSED']
                                     , axis=1)

###################################################################
## Combine "top_10_Rx_cost_bolton" with "temp_join" (a'la VLOOKUP)
##################################################################

#LookUp on "BNF CODE"

top_10_Rx_cost_per_item_with_total_spend_bolton = pd.merge(top_10_Rx_cost_per_item_bolton, 
                                                           temp_join, how='left',left_on='BNF CODE', 
                                                           right_on='BNF CODE')
top_10_Rx_cost_per_item_with_total_spend_bolton.index=['Most expensive', '2nd most expensive', 
                                                       '3rd', '4th','5th', '6th','7th','8th', 
                                                       '9th','10th']
                                                       
                                                       
#######################
# Results table styled
#######################

top_10_Rx_cost_per_item_with_total_spend_bolton_styled=top_10_Rx_cost_per_item_with_total_spend_bolton.style.format(
    {'ACT COST PER ITEM (£)': '{:,.0f}'.format, 'TOTAL ACTUAL COST (£)': '{:,.0f}'.format})

top_10_Rx_cost_per_item_with_total_spend_bolton_styled

##########################################################################
# Barchart comparing ACT COST PER ITEM (£) with TOTAL ACTUAL COST (£) for 
# top 10 most expensive medications
##########################################################################

#Create tuples from the dataframe "top_10_Rx_cost_per_item_with_total_spend_bolton" for barchart

df_for_tuples_a3ii=top_10_Rx_cost_per_item_with_total_spend_bolton.reset_index(drop=True)

med_name_a3ii=tuple(df_for_tuples_a3ii['BNF NAME'].tolist())
med_cost_per_item_a3ii=tuple(df_for_tuples_a3ii['ACT COST PER ITEM (£)'].tolist())
med_actual_cost_a3ii=tuple(df_for_tuples_a3ii['TOTAL ACTUAL COST (£)'].tolist())

ind = np.arange(len(med_cost_per_item_a3ii))  # x locations for the groups
width = 0.4  # width of the bars

fig, ax = plt.subplots(figsize=(9, 6))

rects1 = ax.bar(ind - width/2,med_cost_per_item_a3ii, width,color='c', label='Cost per item')
rects2 = ax.bar(ind + width/2, med_actual_cost_a3ii, width, color='orange', label='Total actual cost')

plt.xticks(rotation=85)
ax.set_title('Barchart comparing the cost per item vs total actual cost of the 10 most expensive medications prescribed', fontsize=14, fontweight='bold')
ax.set_xticks(ind)
ax.set_xticklabels(med_name_a3ii, fontsize=11)
ax.legend()

##########################
# Format X & Y tick values
##########################

fmt = '£{x:,.0f}'

tick = mtick.StrMethodFormatter(fmt)
ax.yaxis.set_major_formatter(tick) 

plt.show()

Assignment A

A (4) Comparing prescribing frequency and costs in Bolton with Cambridge

##################################################################################
## Getting Cambridge numbers need to re-write but I don't have time for this shit!
##################################################################################

#This is for the total number of prescriptions Summing over "ITEMS" column
df_total_number_precriptions_cambridge=pd.DataFrame(df_GP_practice_prescribing_cambridge.groupby('PRACTICE')['ITEMS'].sum())
df_total_number_precriptions_cambridge.columns=['TOTAL NUMBER OF PRESCRIPTIONS']
df_total_number_precriptions_cambridge.reset_index(inplace=True)


#This is for the total cost of prescriptions Summing over "ACT COST" column
df_total_cost_prescritpions_cambridge=pd.DataFrame(df_GP_practice_prescribing_cambridge.groupby('PRACTICE')['ACT COST (£)'].sum())
df_total_cost_prescritpions_cambridge.columns=['TOTAL COST OF PRESCRIPTIONS (£)']
df_total_cost_prescritpions_cambridge.reset_index(inplace=True)

#Combining the total number & total cost of prescriptions

temp_number_cost_prescriptions=pd.merge(df_total_number_precriptions_cambridge,df_total_cost_prescritpions_cambridge)


#A DataFrame with only "PRACTICE" & "PRACTICE NAME" from cambridge DataFrame

temp_presc=df_GP_practice_prescribing_cambridge.loc[:,'PRACTICE':'PRACTICE NAME':8]

#combine temp_number_cost_prescriptions and temp_presc

temp_number_cost_prescriptions_temp_presc=pd.merge(temp_presc,temp_number_cost_prescriptions).drop_duplicates(subset=None, keep='first', inplace=False)

#reindex the above table

QA_4=temp_number_cost_prescriptions_temp_presc.reset_index(drop=True)

##############################################################
## Creating Lists for dataframes
###############################################################

bolton_snapshot=[df_QA_1['TOTAL COST OF PRESCRIPTIONS (£)'].sum(), 
                 df_QA_1['TOTAL NUMBER OF PRESCRIPTIONS'].sum()]


cambridge_snapshot=[QA_4['TOTAL COST OF PRESCRIPTIONS (£)'].sum(), 
                    QA_4['TOTAL NUMBER OF PRESCRIPTIONS'].sum()]

bolton_vs_cambridge_snapshot = pd.DataFrame([bolton_snapshot,cambridge_snapshot],
                 columns=['TOTAL MONTHLY SPEND ON PRESCRIPTIONS (£)','TOTAL NUMBER OF PRESCRIPTIONS'], 
                   index=['Bolton','Cambridge'])


#######################
# Results table styled
#######################

bolton_vs_cambridge_snapshot_styled=bolton_vs_cambridge_snapshot.style.format(
    {'TOTAL MONTHLY SPEND ON PRESCRIPTIONS (£)': '{:,.0f}'.format, 'TOTAL NUMBER OF PRESCRIPTIONS': '{:,.0f}'.format})

bolton_vs_cambridge_snapshot_styled

##########################################################################
# Barchart comparing Bolton & Cambridge prescribing frequencies and spend
##########################################################################

#Create tuples from the dataframe "bolton_vs_cambridge_snapshot" for barchart

bolton= bolton_vs_cambridge_snapshot[0:1].apply(tuple, axis=1).tolist()[0] # 1st row as tuple
cambridge = bolton_vs_cambridge_snapshot[1:2].apply(tuple, axis=1).tolist()[0] # 2nd row as tuple

ind = np.arange(len(bolton))  # x locations for the groups
width = 0.4  # width of the bars

fig, ax = plt.subplots(figsize=(9, 6))

rects1 = ax.bar(ind - width/2,bolton, width,color='orange', label='Bolton')
rects2 = ax.bar(ind + width/2, cambridge, width, color='royalblue', label='Cambridge')

ax.set_title('Barchart comparing Bolton vs Cambridge prescribing frequencies & spend', fontsize=14, fontweight='bold')
ax.set_xticks(ind)
ax.set_xticklabels(('Month spend on prescriptions (£)', 'Number of prescriptions'), fontsize=12, fontweight='bold')
ax.legend()

##########################
# Format X & Y tick values
##########################

fmt = '{x:,.0f}'

tick = mtick.StrMethodFormatter(fmt)
ax.yaxis.set_major_formatter(tick) 

plt.show()

Assignment A

A (5) Using SQL to produce a table with the number of GP practices per city, ordered in descending order.

Note: Cities included are only those mentioned in the coursework.

######################
# SQL for Bolton Only
######################

pdsql = PandaSQL()

sql_query_bolton="""Select count(*) as NUMBER_OF_GP_PRACTICES
             from df_GP_address_ORIG 
              WHERE (POSTCODE LIKE 'BL1%' 
              OR POSTCODE LIKE 'BL2%'OR POSTCODE LIKE 'BL3%' 
              OR POSTCODE LIKE 'BL4%'OR POSTCODE LIKE 'BL5%' 
              OR POSTCODE LIKE 'BL6%'OR POSTCODE LIKE 'BL7%')
              """

df_nbr_gp_bolton=pdsql(sql_query_bolton)
df_nbr_gp_bolton=df_nbr_gp_bolton.rename(index={0: 'BOLTON'})
df_nbr_gp_bolton.head()

#Set up a DataFrame using a subset of the original data.

df_GP_cities_ORIG = pd.read_csv(url_GP_address, names=col_heads_GP_address, 
                                index_col=False, usecols=['POSTCODE'])

#Remove trailing spaces from data.

df_GP_cities_ORIG = df_GP_cities_ORIG.applymap(lambda x: str(x).strip())
data_frame_trimmed = df_GP_cities_ORIG.applymap(lambda x: str(x).strip() if len(str(x).strip()) else None)

##############################
#short postcodes
###############################

test= pd.DataFrame(data_frame_trimmed['POSTCODE'].astype(str).str[:4]).rename(columns = {'POSTCODE':'PRACTICE_SHORT_POSTCODE'})

df_GP_city_postcodes=test.applymap(lambda x: str(x).strip() if len(str(x).strip()) else None)


##################################################################
# Postcodes of each "city" assigned/ mentioned in this coursework 
#################################################################

birmingham_postcodes = ['B1','B10','B11','B12','B13','B14','B15','B16','B17','B18','B19','B20','B21','B23','B24',
'B25','B26','B27','B28','B29','B30','B31','B32','B33','B34','B35','B36','B37','B38','B4','B42',
'B43','B45','B44','B46','B47','B49','B5','B50','B6','B60','B61','B62','B64','B65','B66','B68',
'B67','B69','B7','B70','B71','B72','B73','B74','B75','B76','B78','B77','B79','B8','B80','B9']

bolton_postcodes = ['BL1','BL2','BL3','BL4','BL5','BL6','BL7']

bournemouth_postcodes = ['BH1','BH2','BH3','BH4','BH5','BH6','BH7','BH8','BH9','BH10','BH11']

bradford_postcodes = ['BD1','BD2','BD3','BD4','BD5','BD6','BD7','BD8','BD9','BD10','BD11','BD12',
                      'BD13','BD14','BD15']

brighton_postcodes = ['BN1','BN2','BN41','BN42','BN45','BN50','BN51','BN88']

bristol_postcodes = ['BS0','BS1','BS10','BS11','BS13','BS14','BS15','BS16','BS2','BS20','BS3','BS30',
                     'BS31','BS32','BS34','BS35','BS36','BS37','BS39','BS4','BS40','BS41','BS48',
                     'BS49','BS5','BS6','BS7','BS8','BS9','BS98']

cambridge_postcodes = ['CB1','CB2','CB3','CB4','CB5','CB21','CB22','CB23','CB24','CB25']

coventry_postcodes = ['CV1','CV2','CV3','CV4','CV5','CV6','CV7','CV8']

doncaster_postcodes = ['DN1','DN2','DN3','DN4','DN5','DN6','DN7','DN8','DN9','DN10','DN11','DN12']

leeds_postcodes =['LS1','LS10','LS11','LS12','LS13','LS14','LS15','LS16','LS17','LS18','LS19',
                  'LS2','LS20','LS25','LS26','LS27','LS3','LS4','LS5','LS6','LS7','LS8',
                  'LS88','LS9','LS98','LS99']

leicester_postcodes = ['LE1','LE2','LE3','LE4','LE5','LE6','LE7','LE8','LE9','LE18','LE19']

liverpool_postcodes = ['L1','L10','L11','L12','L13','L14','L15','L16','L17','L18','L19','L2',
                       'L20','L21','L22','L23','L24','L25','L26','L27','L28','L29','L3','L31',
                       'L32','L33','L36','L37','L38','L4','L5','L6','L67','L68','L69','L7','L70',
                       'L71','L72','L73','L74','L75','L8','L9']

luton_postcodes = ['LU1','LU2','LU3','LU4']

manchester_postcodes = ['M1','M11','M12','M13','M14','M15','M16','M17','M18','M19','M2','M20','M21',
                        'M22','M23','M24','M25','M26','M27','M28','M29','M3','M30','M31','M32','M34',
                        'M35','M38','M4','M40','M41','M43','M44','M45','M46','M60','M60','M60','M60',
                        'M60','M61','M8','M9','M90','M99']

nottingham_postcodes = ['NG1','NG10','NG11','NG12','NG13','NG14','NG15','NG16','NG2','NG3','NG4',
                        'NG5','NG6','NG7','NG8','NG80','NG9','NG90','NG17']

plymouth_postcodes = ['PL1','PL2','PL3','PL4','PL5','PL6','PL7','PL8','PL9','PL95']

portsmouth_postcodes = ['PO1','PO2','PO3','PO4','PO5','PO6','PO7','PO8']

sheffield_postcodes = ['S1','S10','S11','S12','S13','S14','S17','S2','S20','S21','S25','S26','S3',
                       'S35','S36','S4','S5','S6','S7','S8','S9','S95','S96','S98','S99']

southampton_postcodes = ['SO14','SO15','SO16','SO17','SO18','SO19']

wolverhampton_postcodes = ['WV1','WV10','WV11','WV2','WV3','WV4','WV5','WV6','WV7','WV8','WV9','WV98','WV99']



postcode_list=birmingham_postcodes+bolton_postcodes+bournemouth_postcodes+bradford_postcodes+brighton_postcodes+bristol_postcodes+cambridge_postcodes+coventry_postcodes+doncaster_postcodes+leeds_postcodes+leicester_postcodes+liverpool_postcodes+luton_postcodes+manchester_postcodes+nottingham_postcodes+plymouth_postcodes+portsmouth_postcodes+sheffield_postcodes+southampton_postcodes+wolverhampton_postcodes

#########################################
# List of cities mentioned in coursework
##########################################

cities_list=["BIRMINGHAM"]*len(birmingham_postcodes) +["BOLTON"]*len(bolton_postcodes) +["BOURNEMOUTH"]*len(bournemouth_postcodes) +["BRADFORD"]*len(bradford_postcodes) +["BRIGHTON"]*len(brighton_postcodes) +["BRISTOL"]*len(bristol_postcodes) +["CAMBRIDGE"]*len(cambridge_postcodes) +["COVENTRY"]*len(coventry_postcodes) +["DONCASTER"]*len(doncaster_postcodes) +["LEEDS"]*len(leeds_postcodes) +["LEICESTER"]*len(leicester_postcodes) +["LIVERPOOL"]*len(liverpool_postcodes) +["LUTON"]*len(luton_postcodes) +["MANCHESTER"]*len(manchester_postcodes) +["NOTTINGHAM"]*len(nottingham_postcodes) +["PLYMOUTH"]*len(plymouth_postcodes) +["PORTSMOUTH"]*len(portsmouth_postcodes) +["SHEFFIELD"]*len(sheffield_postcodes) +["SOUTHAMPTON"]*len(southampton_postcodes) +["WOLVERHAMPTON"]*len(wolverhampton_postcodes)

###################################################
# A datframe of postcodes with corresponding cities
####################################################

df_cities_postcodes=pd.DataFrame({'POSTCODE': postcode_list, 'CITIES': cities_list})

#df_cities_postcodes.head(3)

#####################################################################
#SQL stuff: Join GP practice short postcodes with corresponding city
#####################################################################

sql_query="""SELECT PRACTICE_SHORT_POSTCODE,CITIES
             FROM df_GP_city_postcodes a
             LEFT OUTER JOIN df_cities_postcodes b 
             ON a.PRACTICE_SHORT_POSTCODE = b.POSTCODE
            
         """
c=pdsql(sql_query)

df_city_of_GP_practice=c.dropna()

#df_city_of_GP_practice.head(3)

##################################
#SQL stuff: Number of GP per city
##################################

#df_city_of_GP_practice

#CITIES

sql_query_2="""select CITIES, count(1) as NUMBER_OF_GP_PRACTICES
               from df_city_of_GP_practice
                group by CITIES 
                 order by NUMBER_OF_GP_PRACTICES desc"""
d=pdsql(sql_query_2)

df_nbr_GP_practices_by_city=d.dropna()

df_nbr_GP_practices_by_city.head(20)

#############################################################
# Barchart showing the number og GP practices in each city
###########################################################

#Create tuples from the dataframe "df_nbr_GP_practices_by_city" for barchart

df_for_tuples_a5=df_nbr_GP_practices_by_city.reset_index(drop=True)

cities_a5=tuple(df_for_tuples_a5['CITIES'].tolist())
nbr_gp_prc_a5=tuple(df_for_tuples_a5['NUMBER_OF_GP_PRACTICES'].tolist())


ind = np.arange(len(nbr_gp_prc_a5))  # x locations for the groups
width = 0.6  # width of the bars

fig, ax = plt.subplots(figsize=(9, 6))

rects1 = ax.bar(ind,nbr_gp_prc_a5, width,color='darkkhaki', label='Number of GP Practices')


plt.xticks(rotation=85)
ax.set_title('Barchart showing the number of GP practices in each of the cities mentioned in the assignment', fontsize=14, fontweight='bold')
ax.set_xticks(ind)
ax.set_xticklabels(cities_a5, fontsize=11)


##########################
# Format X & Y tick values
##########################

fmt = '{x:,.0f}'

tick = mtick.StrMethodFormatter(fmt)
ax.yaxis.set_major_formatter(tick) 

plt.show()

Assignment B

B (1) Calculating the monthly total spending for each GP-practice.

#Sum the "ACT COST (£)" column grouped by "PRACTICE"

df_monthly_GP_practice_spend = pd.DataFrame(df_GP_prescribing_ORIG.groupby(df_GP_prescribing_ORIG.PRACTICE).apply(lambda subf: subf['ACT COST (£)'].sum()))


df_monthly_GP_practice_spend.columns = ['MONTHLY PRACTICE SPEND (£)'] # Name the column

df_monthly_GP_practice_spend.reset_index(inplace=True) # Reseting the index


#######################
# Results table styled
#######################

df_monthly_GP_practice_spend_styled=df_monthly_GP_practice_spend.style.format({
    'MONTHLY PRACTICE SPEND (£)': '{:,.0f}'.format})

df_monthly_GP_practice_spend_styled

Assignment B

B (2) Calculating the relative costs per patient

df_GP_spend_merge_with_nbrs_raw = pd.merge(df_monthly_GP_practice_spend, df_GP_patient_nbrs_ORIG, 
                                             how='left',left_on='PRACTICE', right_on='PRACTICE')

df_GP_spend_merge_with_nbrs_raw['COST PER PATIENT (£)'] = df_GP_spend_merge_with_nbrs_raw['MONTHLY PRACTICE SPEND (£)']/df_GP_spend_merge_with_nbrs_raw['TOTAL NUMBER OF PATIENTS']

df_GP_spend_merge_with_nbrs=df_GP_spend_merge_with_nbrs_raw.dropna() 

##############################
# Output Results table styled
#############################

df_GP_spend_merge_with_nbrs_display=df_GP_spend_merge_with_nbrs.style.format({'MONTHLY PRACTICE SPEND (£)': '{:,.0f}'.format,
                                                                              'TOTAL NUMBER OF PATIENTS': '{:,.0f}'.format,
                                                                              'COST PER PATIENT (£)': '{:,.2f}'.format})
df_GP_spend_merge_with_nbrs_display

Assignment B

B (3) Scatterplot with a trend line for GP practices' monthly spending per registered patients

(comparing Bolton practices with nation GP practices)

#################################################################
# Work Out monthly spend per patient for Bolton GP practices only
#################################################################

#Sum the "ACT COST (£)" column grouped by "PRACTICE"

df_monthly_bolton_practice_spend = pd.DataFrame(df_GP_practice_prescribing_bolton.groupby(df_GP_practice_prescribing_bolton.PRACTICE).apply(lambda subf: subf['ACT COST (£)'].sum()))


df_monthly_bolton_practice_spend.columns = ['MONTHLY PRACTICE SPEND (£)'] # Name the column

df_monthly_bolton_practice_spend.reset_index(inplace=True) # Reseting the index


df_bolton_spend_merge_with_nbrs_raw = pd.merge(df_monthly_bolton_practice_spend, df_GP_patient_nbrs_ORIG, 
                                             how='left',left_on='PRACTICE', right_on='PRACTICE')

df_bolton_spend_merge_with_nbrs_raw['COST PER PATIENT (£)'] = df_bolton_spend_merge_with_nbrs_raw['MONTHLY PRACTICE SPEND (£)']/df_bolton_spend_merge_with_nbrs_raw['TOTAL NUMBER OF PATIENTS']

df_bolton_spend_merge_with_nbrs=df_bolton_spend_merge_with_nbrs_raw.dropna() 
#df_bolton_spend_merge_with_nbrs.head(2)

fig=plt.figure(figsize=(14, 8))

ax=fig.add_subplot(111)

x1=df_GP_spend_merge_with_nbrs['TOTAL NUMBER OF PATIENTS']
y1=df_GP_spend_merge_with_nbrs['COST PER PATIENT (£)']

bz1=df_bolton_spend_merge_with_nbrs['TOTAL NUMBER OF PATIENTS']
bz2=df_bolton_spend_merge_with_nbrs['COST PER PATIENT (£)']


#####################################
#Remove outliers for the trendline
#####################################

min_yt = floor(y1.quantile(.01))
max_yt = floor(y1.quantile(.99))

df_pat_nbr_spend = df_GP_spend_merge_with_nbrs.loc[:, ['TOTAL NUMBER OF PATIENTS','COST PER PATIENT (£)']]
df_no_outliers=df_pat_nbr_spend[(df_pat_nbr_spend['COST PER PATIENT (£)']>  min_yt) & (df_pat_nbr_spend['COST PER PATIENT (£)']<  max_yt)]

x1t=df_no_outliers['TOTAL NUMBER OF PATIENTS']
y1t=df_no_outliers['COST PER PATIENT (£)']

############
#Trend line
############

fit = np.polyfit(x1t, y1t, deg=1)

ax.plot(x1t, fit[0] * x1t + fit[1], color='red', alpha=1,label='Trend Line')


######################################
# Scatter Plot for all GP patient nbrs
######################################

ax.scatter(x1, y1, color='royalblue', label='All GP Practices')

#########################################
# Scatter Plot for Bolton GP patient nbrs 
#########################################

ax.scatter(bz1, bz2, color='orange',label='Bolton GP Practices')


ax.set_title('Scatterplot showing the number of registered patients vs monthly per patient spend (January 2016)', 
             fontsize=15, fontweight='bold')
ax.set_xlabel("Number of Registered Patients",fontsize=14)
ax.set_ylabel("Monthly Spend (per patient)",fontsize=14)


max_x = floor(x1.max())+5000
max_y = floor(bz2.max())+10

ax.set_xlim(0, max_x)
ax.set_ylim(0, max_y)

##########################
# Format X & Y tick values
##########################

fmt_pound = '£{x:,.0f}'
fmt_k= '{x:,.0f}'

tick_pound = mtick.StrMethodFormatter(fmt_pound)
tick_k = mtick.StrMethodFormatter(fmt_k)

ax.xaxis.set_major_formatter(tick_k) 
ax.yaxis.set_major_formatter(tick_pound) 

min_x1 = floor(x1.quantile(.01))
max_x1 = floor(x1.quantile(.99))
plt.xlim(min_x1, max_x1) #do not show extreme outliers

plt.legend(loc='best');

Assignment B

B (4) Histogram showing national GP practices' monthly spending per registered patients

fig=plt.figure(figsize=(8, 5))

x2=df_GP_spend_merge_with_nbrs['COST PER PATIENT (£)']
x2t=df_no_outliers['COST PER PATIENT (£)']

plt.hist(x2t, bins=50, color='lightgreen')

plt.title("Histogram showing GP practices' monthly prescription spend (per registered patient)",fontsize=15, fontweight='bold')
plt.xlabel("Monthly Spend (£)",fontsize=12)
plt.ylabel('Frequency',fontsize=12);


min_x2 = floor(x2.quantile(.01))
max_x2 = floor(x2.quantile(.99))
plt.xlim(min_x2, max_x2) #do not show extreme outliers
plt.show()

Assignment B

B (5) Comparing Bolton GP practices to the national level using descriptive statistics

############################################################
# Descriptive statistics for GP Practices' spend per patient
############################################################

temp_nation=df_GP_spend_merge_with_nbrs ['COST PER PATIENT (£)'].describe()
temp_bolton=df_bolton_spend_merge_with_nbrs ['COST PER PATIENT (£)'].describe()


df_temp_nation = pd.DataFrame([temp_nation], index=['National Level'])
df_temp_bolton = pd.DataFrame([temp_bolton], index=['Bolton GP Practices'])

df_per_patient_spend_descriptive_stats=pd.concat([df_temp_nation,df_temp_bolton])


pd.options.display.float_format = '{:,.2f}'.format #for display


df_per_patient_spend_descriptive_stats.head()

#df_per_patient_spend_descriptive_stats.dtypes

#################################################
# Descriptive statistics for GP Practices' monthly spend 
#################################################

temp_nation2=df_GP_spend_merge_with_nbrs_raw ['MONTHLY PRACTICE SPEND (£)'].describe()
temp_bolton2=df_bolton_spend_merge_with_nbrs_raw ['MONTHLY PRACTICE SPEND (£)'].describe()


df_temp_nation2 = pd.DataFrame([temp_nation2], index=['National Level'])
df_temp_bolton2 = pd.DataFrame([temp_bolton2], index=['Bolton GP Practices'])

df_monthly_spend_descriptive_stats=pd.concat([df_temp_nation2,df_temp_bolton2])


pd.options.display.float_format = '{:,.0f}'.format #for display

df_monthly_spend_descriptive_stats.head()

#df_per_patient_spend_descriptive_stats.dtypes

##################################################
# Boxplots for for GP Practices' spend per patient
##################################################

#Need the following in order to avoid warnings when trying to use column values

temp_nation3 = df_GP_spend_merge_with_nbrs ['COST PER PATIENT (£)']
temp_nation_np3 = np.array(temp_nation3)
temp_nation_np3 = temp_nation_np3.reshape([temp_nation3.shape[0], 1]) 

temp_bolton3 = df_bolton_spend_merge_with_nbrs ['COST PER PATIENT (£)']
temp_bolton_np3 = np.array(temp_bolton3)
temp_bolton_np3 = temp_bolton_np3.reshape([temp_bolton3.shape[0], 1]) 



fig=plt.figure(figsize=(8, 5))

fig.suptitle('Boxplots showing national vs Bolton GP practices spend per patient', fontsize=15, fontweight='bold')

#plt.boxplot([df_GP_spend_merge_with_nbrs ['COST PER PATIENT (£)'],
             #df_bolton_spend_merge_with_nbrs ['COST PER PATIENT (£)']], 0, 'g', 0) This code gives warnings!

plt.boxplot([temp_nation_np3,temp_bolton_np3], 0, 'g', 0)
    
plt.xlabel('GP Practice Spend - per patient (£)', fontsize=12, fontweight='bold')
plt.yticks([1,2],['National Level', 'Bolton'], fontsize=12, fontweight='bold')

min_x3 = floor(df_GP_spend_merge_with_nbrs ['COST PER PATIENT (£)'].quantile(.01))
max_x3 = floor(df_GP_spend_merge_with_nbrs ['COST PER PATIENT (£)'].quantile(.99))
plt.xlim(min_x3, max_x3) #do not show extreme outliers

plt.show()

##################################################
# Boxplots for for GP Practices' spend per month
##################################################

#Need the following in order to avoid warnings when trying to use column values

temp_nation4 = df_GP_spend_merge_with_nbrs_raw ['MONTHLY PRACTICE SPEND (£)']
temp_nation_np4 = np.array(temp_nation4)
temp_nation_np4 = temp_nation_np4.reshape([temp_nation4.shape[0], 1]) 

temp_bolton4 = df_bolton_spend_merge_with_nbrs_raw ['MONTHLY PRACTICE SPEND (£)']
temp_bolton_np4 = np.array(temp_bolton4)
temp_bolton_np4 = temp_bolton_np4.reshape([temp_bolton4.shape[0], 1]) 



fig=plt.figure(figsize=(8, 5))

fig.suptitle('Boxplots showing national vs Bolton GP practices monthly spend', fontsize=15, fontweight='bold')


plt.boxplot([temp_nation_np4,temp_bolton_np4], 0, 'g', 0)
    
plt.xlabel('GP Practice Monthly Spend (£)', fontsize=12, fontweight='bold')
plt.yticks([1,2],['National Level', 'Bolton'], fontsize=12, fontweight='bold')

min_x4 = floor(df_GP_spend_merge_with_nbrs_raw ['MONTHLY PRACTICE SPEND (£)'].quantile(.01))-10000
max_x4 = floor(df_GP_spend_merge_with_nbrs_raw ['MONTHLY PRACTICE SPEND (£)'].quantile(.99))
plt.xlim(min_x4, max_x4) #do not show extreme outliers

plt.show()

Assignment C

Read in Data

ONS Postcode Data

#####################################################################
## Read in the "ONS Postcode Lookup" data & setup a pandas dataframe
#####################################################################

# Store the url of the data as a variable
url_ons_postcodes = 'https://s3.eu-west-2.amazonaws.com/dmhr-data/postcodes.csv'

#Set up a DataFrame using a subset of the original data.
df_ons_postcodes_ORIG= pd.read_csv(url_ons_postcodes, index_col=False, 
                                   usecols=['Postcode 1','Postcode 2', 'Postcode 3','Region Name'])

#Rename column headings
df_ons_postcodes_ORIG.rename(columns={'Postcode 1':'POSTCODE 1','Postcode 2':'POSTCODE 2', 
                                      'Postcode 3':'POSTCODE 3','Region Name':'REGION NAME'}, 
                             inplace=True)


#Remove trailing spaces from data.
#df_ons_postcodes_ORIG = df_ons_postcodes_ORIG.applymap(lambda x: str(x).strip() if len(str(x).strip()) else None)

df_ons_postcodes_ORIG.head(1)

Read in Data

Deprivation Indices Data

############################################################################
## Read in the "English deprivation indices" data & setup a pandas dataframe
############################################################################

# Store the url of the data as variables
url_eng_deprivation_indices = 'https://s3.eu-west-2.amazonaws.com/dmhr-data/deprivation-by-postcode.csv'

#Set up a DataFrame using a subset of the original data.
df_eng_deprivation_indices_ORIG= pd.read_csv(url_eng_deprivation_indices, 
                                             index_col=False, 
                                             usecols=['Postcode','Postcode Status',
                                                      'LSOA code','LSOA Name',
                                                      'Index of Multiple Deprivation Rank',
                                                      'Index of Multiple Deprivation Decile'])
#Rename column headings
df_eng_deprivation_indices_ORIG.rename(columns={'Postcode':'POSTCODE',
                                                'Postcode Status':'POSTCODE STATUS',
                                                'LSOA code':'LSOA CODE',
                                                'LSOA Name':'LSOA NAME',
                                                'Index of Multiple Deprivation Rank':'IMD RANK',
                                                'Index of Multiple Deprivation Decile':'IMD DECILE'}, 
                                       inplace=True)

df_eng_deprivation_indices_ORIG = df_eng_deprivation_indices_ORIG.dropna() 

#Remove trailing spaces from data.
df_eng_deprivation_indices_ORIG = df_eng_deprivation_indices_ORIG.applymap(lambda x: str(x).strip())

# Drop rows for which no entries in csv file this for 10 postcodes that are **UNMATCHED**
df_eng_deprivation_indices_ORIG = df_eng_deprivation_indices_ORIG.dropna() 

#Reformat columns that were converted into Objects in the process of removing white space
#They get formatted into float64

df_eng_deprivation_indices_ORIG['IMD RANK'] = pd.to_numeric(df_eng_deprivation_indices_ORIG['IMD RANK'])
df_eng_deprivation_indices_ORIG['IMD DECILE'] = pd.to_numeric(df_eng_deprivation_indices_ORIG['IMD DECILE'])

df_eng_deprivation_indices_ORIG.head(1)

Read in Data

CVD Mortality Rates Data

#############################################################################
## Read in the "Under 75 CVD mortality rates" data & setup a pandas dataframe
#############################################################################

# Store the url of the data as variables

url_mort_rates = 'https://s3.eu-west-2.amazonaws.com/dmhr-data/NHSOF_1.1_I00656_D.csv'

#Set up a DataFrame using all original data.

df_mort_rates_ORIG= pd.read_csv(url_mort_rates)

#Rename column headings

df_mort_rates_ORIG.rename(columns={'Year':'YEAR','Period of coverage':'PERIOD OF COVERAGE',
                                   'Breakdown':'BREAKDOWN','Level':'LEVEL',
                                   'Level description':'LEVEL DESCRIPTION','Gender':'GENDER',
                                   'Age':'AGE','Indicator value':'INDICATOR VALUE',
                                   'Lower CI':'LOWER CI','Upper CI':'UPPER CI',
                                   'Numerator':'NUMERATOR','Denominator':'DENOMINATOR'}, 
                          inplace=True)


#Remove trailing spaces from data.

df_mort_rates_ORIG = df_mort_rates_ORIG.applymap(lambda x: str(x).strip())

#Reformat columns that were converted into Objects in the process of removing white space
#They get formatted into int64 an float64

df_mort_rates_ORIG['YEAR'] = pd.to_numeric(df_mort_rates_ORIG['YEAR']) 

df_mort_rates_ORIG['INDICATOR VALUE'] = pd.to_numeric(df_mort_rates_ORIG['INDICATOR VALUE'].str.replace(r'[,]','')).astype(np.float64) 
df_mort_rates_ORIG['LOWER CI'] = pd.to_numeric(df_mort_rates_ORIG['LOWER CI'].str.replace(r'[,]','')).astype(np.float64)
df_mort_rates_ORIG['UPPER CI'] = pd.to_numeric(df_mort_rates_ORIG['UPPER CI'].str.replace(r'[,]','')).astype(np.float64)
df_mort_rates_ORIG['NUMERATOR'] = pd.to_numeric(df_mort_rates_ORIG['NUMERATOR'].str.replace(r'[,]','')).astype(np.float64)
df_mort_rates_ORIG['DENOMINATOR'] = pd.to_numeric(df_mort_rates_ORIG['DENOMINATOR'].str.replace(r'[,]','')).astype(np.float64)
 
df_mort_rates_ORIG.head(1)

Assignment C

C (1) The relative costs per patient for all statin prescriptions [simvastatin, atorvastatin, rosuvastatin, pravastatin, fluvastatin] for all GP-practices [December 2015 dataset]

################################
# A list of statin prescriptons
################################

statin_list='Simvastatin|Atorvastatin|Rosuvastatin|Pravastatin|Fluvastatin'

###########################################################################
# Modify "df_GP_prescribing_ORIG" to consider only prescriptions of statins
###########################################################################


df_GP_prescribing_statins = df_GP_prescribing_ORIG[df_GP_prescribing_ORIG['BNF NAME'].str.contains(statin_list, case=False)]

#df_GP_prescribing_statins.head(3)

#This is for the total cost of statin prescriptions Summing over "ACT COST" column

df_total_cost_statin_prescritpions=pd.DataFrame(df_GP_prescribing_statins.groupby('PRACTICE')['ACT COST (£)'].sum())
df_total_cost_statin_prescritpions.columns=['TOTAL COST OF STATIN PRESCRIPTIONS (£)']
df_total_cost_statin_prescritpions.reset_index(inplace=True)

#df_total_cost_statin_prescritpions.head()

#df_total_cost_statin_prescritpions['TOTAL COST OF STATIN PRESCRIPTIONS (£)'].sum()

######################################################################################################
# Combine "df_total_cost_statin_prescritpions" with number of patients with "df_GP_patient_nbrs_ORIG"
######################################################################################################

df_statin_patient_nbr_merged_raw = pd.merge(df_GP_patient_nbrs_ORIG, df_total_cost_statin_prescritpions, 
                                        how='left',left_on='PRACTICE', right_on='PRACTICE')

#Drop Null Rows

df_statin_patient_nbr_merged=df_statin_patient_nbr_merged_raw.dropna()

#############################################################################################
# Getting warnning when trying df['new column name'] = df['column name1']/df['column name2']
#so have to go from first principles as below
#############################################################################################

df = pd.DataFrame(df_statin_patient_nbr_merged['TOTAL COST OF STATIN PRESCRIPTIONS (£)']/df_statin_patient_nbr_merged['TOTAL NUMBER OF PATIENTS'],
                 columns = ['STATIN COST PER PATIENT (£)'])     

##############################################################
# Relative costs per patient for all statin prescriptions 
# Taking the average over all patients registered at practice
##############################################################

df_statin_cost_per_patient= df_statin_patient_nbr_merged.join(df)


#######################
# Results table styled
#######################

df_statin_cost_per_patient_styled=df_statin_cost_per_patient.style.format({
    'TOTAL NUMBER OF PATIENTS': '{:,.0f}'.format,
    'TOTAL COST OF STATIN PRESCRIPTIONS (£)': '{:,.0f}'.format,'STATIN COST PER PATIENT (£)': '{:,.2f}'.format})

df_statin_cost_per_patient_styled

Assignment C

C (2) The associated Index of Multiple Deprivation (IMD) for each GP-Practice in Bolton

################################################################
# Modify "df_GP_address_ORIG" to consider only Bolton postcodes
################################################################

df_GP_address_bolton_postcodes = df_GP_address_ORIG[df_GP_address_ORIG['POSTCODE'].str.contains('BL1|BL2|BL3|BL4|BL5|BL6|BL7', case=False)]

#df_GP_address_bolton_postcodes.head(2)

######################################################################
# Modify the IMD table to keep only relevant columns for this analysis
######################################################################

df_eng_deprivation_indices=df_eng_deprivation_indices_ORIG.drop(['POSTCODE STATUS','LSOA CODE','LSOA NAME'], axis=1)

#df_eng_deprivation_indices.head(2)

################################################################################
# Associated Index of Multiple Deprivation (IMD) for each GP-Practice in England.
# Merging "df_eng_deprivation_indices" and "df_GP_address_ORIG"
################################################################################

df_eng_GP_deprivation_indices=pd.merge(df_GP_address_ORIG, df_eng_deprivation_indices, 
                                       how='left',left_on='POSTCODE', right_on='POSTCODE')

df_eng_GP_deprivation_indices=df_eng_GP_deprivation_indices.drop_duplicates(subset=['PRACTICE'], keep='first')

#######################
# Drop all Null Values
#######################

df_eng_GP_deprivation_indices_raw=df_eng_GP_deprivation_indices.dropna()

df_eng_GP_deprivation_indices=df_eng_GP_deprivation_indices_raw.reset_index(drop=True)

################################################################################
# Associated Index of Multiple Deprivation (IMD) for each GP-Practice in Bolton.
# Merging "df_eng_deprivation_indices" and "df_GP_address_bolton_postcodes"
################################################################################

df_bolton_deprivation_indices_raw=pd.merge(df_GP_address_bolton_postcodes, df_eng_deprivation_indices, 
                                       how='left',left_on='POSTCODE', right_on='POSTCODE')

df_bolton_deprivation_indices_raw=df_bolton_deprivation_indices_raw.drop_duplicates(subset=['PRACTICE'], keep='first')

df_bolton_deprivation_indices=df_bolton_deprivation_indices_raw.reset_index(drop=True)

#######################
# Results table styled
#######################

df_bolton_deprivation_indices_styled=df_bolton_deprivation_indices.style.format({
    'IMD RANK': '{:,.0f}'.format,
    'IMD DECILE': '{:,.0f}'.format})

df_bolton_deprivation_indices_styled

Assignment C

C (3) i. The lowest relative spenders on statins from the first decile and the highest relative spenders on statins from the last decile from the entire national dataset

#########################################################
# Place values of relative spend on statins into deciles
#########################################################


df_statin_cost_per_patient['Decile'] = pd.qcut(df_statin_cost_per_patient['STATIN COST PER PATIENT (£)'],
                                                                    10, labels=False)
#df_statin_cost_per_patient.head()

#######################################################
# Lowest relative spenders on statins from first decile
#######################################################

df_statin_cost_per_patient_first_decile = df_statin_cost_per_patient[df_statin_cost_per_patient.Decile== 9]

df_statin_cost_per_patient_first_decile_low_to_high_raw=df_statin_cost_per_patient_first_decile.sort_values(by=['STATIN COST PER PATIENT (£)'], ascending=True)

df_statin_cost_per_patient_first_decile_low_to_high=df_statin_cost_per_patient_first_decile_low_to_high_raw.reset_index(drop=True)

#######################
# Results table styled
#######################

df_statin_cost_per_patient_first_decile_low_to_high_styled=df_statin_cost_per_patient_first_decile_low_to_high.style.format({
    'TOTAL NUMBER OF PATIENTS': '{:,.0f}'.format,
    'TOTAL COST OF STATIN PRESCRIPTIONS (£)': '{:,.0f}'.format,
     'STATIN COST PER PATIENT (£)': '{:,.2f}'.format,
    'Decile': '{:,.0f}'.format})

df_statin_cost_per_patient_first_decile_low_to_high_styled

########################################################
# Highest relative spenders on statins from last decile
########################################################


df_statin_cost_per_patient_last_decile = df_statin_cost_per_patient[df_statin_cost_per_patient.Decile== 0]

df_statin_cost_per_patient_last_decile_high_to_low_raw=df_statin_cost_per_patient_last_decile.sort_values(by=['STATIN COST PER PATIENT (£)'], ascending=False)

df_statin_cost_per_patient_last_decile_high_to_low=df_statin_cost_per_patient_last_decile_high_to_low_raw.reset_index(drop=True)

#######################
# Results table styled
#######################

df_statin_cost_per_patient_last_decile_high_to_low_styled=df_statin_cost_per_patient_last_decile_high_to_low.style.format({
    'TOTAL NUMBER OF PATIENTS': '{:,.0f}'.format,
    'TOTAL COST OF STATIN PRESCRIPTIONS (£)': '{:,.0f}'.format,
     'STATIN COST PER PATIENT (£)': '{:,.2f}'.format,
    'Decile': '{:,.0f}'.format})

df_statin_cost_per_patient_last_decile_high_to_low_styled

Assignment C

C (3) ii. The associated Index of Multiple Deprivation (IMD) for all identified GP-practices for both groups (lowest and the highest), assessing whether the IMD-score differs.

#######################################################
# First decile associated Index of Multiple Deprivation
#######################################################


df_first_decile_statin_spend_and_IMD=pd.merge(df_statin_cost_per_patient_first_decile_low_to_high, 
                                              df_eng_GP_deprivation_indices, 
                                       how='left',left_on='PRACTICE', right_on='PRACTICE')

df_first_decile_statin_spend_and_IMD_all=df_first_decile_statin_spend_and_IMD.drop_duplicates(subset=['PRACTICE'], keep='first')

df_first_decile_statin_spend_and_IMD=df_first_decile_statin_spend_and_IMD_all.drop(['PRACTICE NAME','POSTCODE'], axis=1)

#df_first_decile_statin_spend_and_IMD.head()

#######################################################
# Last decile associated Index of Multiple Deprivation
#######################################################

df_last_decile_statin_spend_and_IMD=pd.merge(df_statin_cost_per_patient_last_decile_high_to_low, 
                                              df_eng_GP_deprivation_indices, 
                                       how='left',left_on='PRACTICE', right_on='PRACTICE')

df_last_decile_statin_spend_and_IMD_all=df_last_decile_statin_spend_and_IMD.drop_duplicates(subset=['PRACTICE'], keep='first')

df_last_decile_statin_spend_and_IMD=df_last_decile_statin_spend_and_IMD_all.drop(['PRACTICE NAME','POSTCODE'], axis=1)

#df_last_decile_statin_spend_and_IMD.head()

##########################################
# Descriptive statistics for IMD scores
##########################################

temp_last=df_last_decile_statin_spend_and_IMD ['IMD RANK'].describe()
temp_first=df_first_decile_statin_spend_and_IMD ['IMD RANK'].describe()


df_temp_first = pd.DataFrame([temp_first], index=['First Decile'])
df_temp_last = pd.DataFrame([temp_last], index=['Last Decile'])


df_IMD_deciles_descriptive_stats=pd.concat([df_temp_first,df_temp_last])

pd.options.display.float_format = '{:,.0f}'.format #for display

df_IMD_deciles_descriptive_stats.head()

####################################################
# Boxplots for IMD scores of first and last deciles
###################################################

#Need the following in order to avoid warnings when trying to use column values

temp_first = df_first_decile_statin_spend_and_IMD['IMD RANK']
temp_first_np = np.array(temp_first)

temp_last = df_last_decile_statin_spend_and_IMD['IMD RANK']
temp_last_np = np.array(temp_last)

fig=plt.figure(figsize=(8, 5))

fig.suptitle('Boxplots of first & last deciles', fontsize=15, fontweight='bold')

plt.boxplot([temp_first_np,temp_last_np], 0, 'g', 0)
    
plt.xlabel('IMD Score', fontsize=12, fontweight='bold')
plt.yticks([1,2],['First decile', 'Last decile'], fontsize=12, fontweight='bold')


plt.show()

Assignment C

C (4) GP-practices & their regions together with the associated 75 mortality rate for cardiovascular diseases for the year 2015 for each region.

###########################################################################
# Here I modify "df_mort_rates_ORIG" to consider only regions & 2015
###########################################################################

mort_year=df_mort_rates_ORIG['YEAR']
mort_breakdown=df_mort_rates_ORIG['BREAKDOWN']
mort_gender=df_mort_rates_ORIG['GENDER']
mort_age=df_mort_rates_ORIG['AGE']

df_mort_rates_Raw= df_mort_rates_ORIG.loc[(mort_year == 2015) & (mort_breakdown == 'Region')
                                       & (mort_gender == 'Person') & (mort_age == 'All')
                                       , ["LEVEL DESCRIPTION","NUMERATOR","DENOMINATOR"]] 
##############################
#
#####################################
series_temp_rates=df_mort_rates_Raw['NUMERATOR']/df_mort_rates_Raw['DENOMINATOR']

df_temp_rates = pd.DataFrame(series_temp_rates, columns = ['MORTALITY RATES'])  

###################################
#
##################################

df_mort_rates_for_Q_temp = pd.concat([df_mort_rates_Raw,df_temp_rates], axis=1)
df_mort_rates_for_Q_temp = df_mort_rates_for_Q_temp.rename(columns={'LEVEL DESCRIPTION': 'REGION'})


#########################################
# Drop Numerator & Denominator Colummns
#########################################

df_mort_rates_for_Q_raw=df_mort_rates_for_Q_temp.drop(['NUMERATOR','DENOMINATOR'], axis=1)

df_mort_rates_for_Q = df_mort_rates_for_Q_raw.reset_index(drop=True)

#######################
# Results table styled
#######################

df_mort_rates_for_Q_styled=df_mort_rates_for_Q.style.format({
    'MORTALITY RATES': '{:,.4%}'.format})

df_mort_rates_for_Q_styled

Assignment C

C (5) A Visualisation (using matplotlib) of each region's spending on statins [x-axis] and the mortality rate [y-axis]. Assessing whether relative spending for statin prescriptions in each regions correlates with the mortality rate from cardiovascular diseases.

################################################################
# Merge "df_statin_patient_nbr_merged" and "df_GP_address_ORIG"
################################################################

df_merge_c5 = pd.merge(df_statin_patient_nbr_merged,df_GP_address_ORIG, how='left',
                                          left_on='PRACTICE', right_on='PRACTICE')

####################################
# Remove blank spaces from postcodes
####################################

series_GP_postcode_no_blanks=df_merge_c5['POSTCODE'].str.replace(' ', '')
df_GP_postcode_no_blanks = pd.DataFrame(series_GP_postcode_no_blanks)  

##################################################
#Drop postcodes and merge postcodes with no blanks
##################################################

df_merge_c5_short=df_merge_c5.drop(['PRACTICE NAME','POSTCODE'], axis=1)

statin_spend_postcodes_no_blanks= pd.concat([df_merge_c5_short, df_GP_postcode_no_blanks], axis = 1)
#statin_spend_postcodes_no_blanks.head()

###########################################################
# Modify "df_ons_postcodes_ORIG" to include mortality rates
###########################################################

df_ons_postcodes_with_mort_rates_raw = pd.merge(df_ons_postcodes_ORIG, df_mort_rates_for_Q, how='left',
                                          left_on='REGION NAME', right_on='REGION')

####################################
# Remove blank spaces from postcodes
####################################

series_ons_postcode_no_blanks=df_ons_postcodes_with_mort_rates_raw['POSTCODE 1'].str.replace(' ', '')

df_ons_postcode_no_blanks = pd.DataFrame(series_ons_postcode_no_blanks)  

##################################################
#Drop postcodes and merge postcodes with no blanks
##################################################

df_ons_postcodes_with_mort_rates_short=df_ons_postcodes_with_mort_rates_raw.drop(['POSTCODE 1','POSTCODE 2',
                                                                                  'POSTCODE 3', 'REGION NAME'], axis=1)

test_c5= pd.concat([df_ons_postcodes_with_mort_rates_short, df_ons_postcode_no_blanks], axis = 1)

test= pd.merge(statin_spend_postcodes_no_blanks, test_c5, how='left',
                                          left_on=['POSTCODE'], right_on=['POSTCODE 1'])

#####################################################################################
# 289 null values out of 764 values due to postcodes from surgery addresses not 
# matching ONS postcodes Will impute values of null matching postcodes using 'ffill'
# The 'ffill' option to 'forward fill' and propagate last valid observation forward
#####################################################################################

df_a=test.sort_values(by=['POSTCODE'])

df_b=df_a.fillna(method='ffill')

df_c = df_b.sort_values(by=['PRACTICE'])

df_d=df_c.drop(['PRACTICE','POSTCODE','POSTCODE','POSTCODE 1'], axis=1)

#This is for the total spend on statin prescriptions 
#Summing over "TOTAL COST OF STATIN PRESCRIPTIONS (£)" column

df_g = df_d.groupby(['REGION','MORTALITY RATES']).sum().reset_index()

#######################################
# Find total regional spend on statins
#######################################

df_n=df_g.drop(['TOTAL NUMBER OF PATIENTS',], axis=1)

##################################
# Find regional spend per patient
#################################

df_h=df_g['TOTAL COST OF STATIN PRESCRIPTIONS (£)']/df_g['TOTAL NUMBER OF PATIENTS']
df_j = pd.DataFrame(df_h, columns = ['REGIONAL STATIN SPEND PER PATIENT'])  

#########################################################################
# Drop TOTAL COST OF STATIN PRESCRIPTIONS (£) & TOTAL NUMBER OF PATIENTS
#########################################################################

df_k=df_g.drop(['TOTAL COST OF STATIN PRESCRIPTIONS (£)','TOTAL NUMBER OF PATIENTS',], axis=1)


df_m= pd.concat([df_k, df_j], axis = 1)

df_m

#######################
# Results table styled
#######################

df_m_styled=df_m.style.format({
    'MORTALITY RATES': '{:,.4%}'.format,
    'REGIONAL STATIN SPEND PER PATIENT': '£{:,.2f}'.format})

df_m_styled

####################################################
# Plot total relative spend on statins vs mortality
####################################################

data=df_m.values # convert dataframe into pandasnumarray

regions_list=data[:,0].tolist() # List of all regions
nbr_regions =len(regions_list) # number of regions

label = list(range(0, nbr_regions)) # list fom 0,1,2....,number of regions

x = data[:,2].tolist() # Region spending on statins as list for x-axis
y = data[:,1].tolist() # Mortality rate as list for y-axis


# Define a colormap with the right number of colors
cmap = plt.cm.get_cmap('jet',max(label)-min(label)+1)

bounds = range(min(label),max(label)+2)
norm = colors.BoundaryNorm(bounds, cmap.N)


plt.figure(figsize=(14, 7)) #Size of scatter plot
axes_bounds = [min(x)-0.05, max(x)+0.05, min(y)-0.00001, max(y)+0.00001] # Limits of x and y-axes

plt.scatter(x, y, c= label, s=100, cmap=cmap, norm=norm) 

# Add a colorbar. Move the ticks up by 0.5, so they are centred on the colour.

cb=plt.colorbar(ticks=np.array(label)+0.5)
cb.set_ticklabels(regions_list)

plt.title('Relative Spend on Statins in England vs Moratlity Rates (by region)', fontsize=14, fontweight='bold')
plt.xlabel("Statin Spend (per patient)",fontsize=12, fontweight='bold') # Label for the x-axis
plt.ylabel("Mortality Rate",fontsize=12, fontweight='bold') # Label for the y-axis
plt.grid(True) #Show Gridlines

plt.axis(axes_bounds) # set bounds of x and y-axes

plt.show()

# Matrix of correlations of x with x, x with y, y with x and y with y. 
# Key column is position (1, 0) or (0, 1).

Corr_matrix_1 = pd.DataFrame(np.corrcoef(x, y))

Corr_matrix_2 = Corr_matrix_1.rename(columns={0: 'MORTALITY RATES', 1: 'STATIN SPEND'})
Corr_matrix = Corr_matrix_2.rename(index={0: 'MORTALITY RATES', 1: 'STATIN SPEND'})

print("""The value of the correlation of mortality rates & statin spend is %s"""% Corr_matrix.iat[0,1]) 
print("""The value of the correlation of statin spend & mortality rates is %s"""% Corr_matrix.iat[1,0])
print("""""")
print("""There is a strong positive correlation. This suggests that the relative spending """)
print("""for statin prescriptions in each region increases when the region has high mortality""")
print("""rates from cardiovascular diseases.""")

The value of the correlation of mortality rates & statin spend is 0.8716714658596678
The value of the correlation of statin spend & mortality rates is 0.8716714658596678

There is a strong positive correlation. This suggests that the relative spending 
for statin prescriptions in each region increases when the region has high mortality
rates from cardiovascular diseases.

#correlation matrix plot

df_z1=df_m.drop(['REGION'], axis=1)

df_z=df_z1.rename(columns={'REGIONAL STATIN SPEND PER PATIENT': 'STATIN SPEND'})

plt.matshow(df_z.corr())
plt.xticks(range(len(df_z.columns)), df_z.columns)
plt.yticks(range(len(df_z.columns)), df_z.columns)
plt.colorbar()
plt.show()

#######################################################
# Plot total regional spend on statins vs mortality
#######################################################

data1=df_n.values # convert dataframe into pandasnumarray

regions_list1=data1[:,0].tolist() # List of all regions
nbr_regions1 =len(regions_list1) # number of regions

label1 = list(range(0, nbr_regions1)) # list fom 0,1,2....,number of regions

x1 = data1[:,2].tolist() # Region spending on statins as list for x-axis
y1 = data1[:,1].tolist() # Mortality rate as list for y-axis


# Define a colormap with the right number of colors
cmap = plt.cm.get_cmap('jet',max(label1)-min(label1)+1)

bounds1 = range(min(label1),max(label1)+2)
norm1 = colors.BoundaryNorm(bounds, cmap.N)


plt.figure(figsize=(14, 7)) #Size of scatter plot
axes_bounds1 = [min(x1)-10000, max(x1)+10000, min(y1)-0.00001, max(y1)+0.00001] # Limits of x and y-axes

plt.scatter(x1, y1, c= label1, s=100, cmap=cmap, norm=norm) 

# Add a colorbar. Move the ticks up by 0.5, so they are centred on the colour.

cb=plt.colorbar(ticks=np.array(label1)+0.5)
cb.set_ticklabels(regions_list1)

plt.title('Total Spend on Statins in England vs Moratlity Rates (by region)', fontsize=14, fontweight='bold')
plt.xlabel("Statin Spend",fontsize=12, fontweight='bold') # Label for the x-axis
plt.ylabel("Mortality Rate",fontsize=12, fontweight='bold') # Label for the y-axis
plt.grid(True) #Show Gridlines

plt.axis(axes_bounds1) # set bounds of x and y-axes

plt.show()

Assignment D

Read in Data: Google FluTrends Surveillance Data

#################################################
## Read in the data files to a pandas dataframe
#################################################

# Store the url of the data as variables

url_romania = 'https://www.google.org/flutrends/about/data/flu/ro/data.txt'
url_uruguay = 'https://www.google.org/flutrends/about/data/flu/uy/data.txt'

# Create suitable column titles

col_heads_romania = ['Date', '# of cases of flu in Romania']
col_heads_uruguay = ['Date', '# of cases of flu in Uruguay']


#Set up DataFrames. Need to set header=8 to properly extract data

df_romania_FluTrends = pd.read_csv(url_romania,header=8,names=col_heads_romania, index_col=False)
df_romania_FluTrends['Date'] = pd.to_datetime(df_romania_FluTrends['Date'])

df_uruguay_FluTrends = pd.read_csv(url_uruguay,header=8,names=col_heads_uruguay, index_col=False)
df_uruguay_FluTrends['Date'] = pd.to_datetime(df_uruguay_FluTrends['Date'])

Assignment D

D (1) A visualisation of the seasonal patterns across all years

#######################################
## A plot of flu trends in Romania only
#######################################

romania_plot=df_romania_FluTrends.plot(legend ='left', x='Date', y = '# of cases of flu in Romania', 
                                       c='r',figsize=(10, 6), grid=True)

romania_plot.set_title('Google FluTrends surveillance data (Romania) - Across all years',fontsize=15, fontweight='bold')
romania_plot.set_ylabel('Frequency (cases of flu)',fontsize=14)
romania_plot.set_xlabel('Year',fontsize=14)
plt.show()

#######################################
## A plot of flu trends in Uruguay only
#######################################

uruguay_plot=df_uruguay_FluTrends.plot(legend ='left', x='Date', y = '# of cases of flu in Uruguay', figsize=(10, 6), grid=True)

uruguay_plot.set_title('Google FluTrends surveillance data (Uruguay) - Across all years',fontsize=15, fontweight='bold')
uruguay_plot.set_ylabel('Frequency (cases of flu)',fontsize=14)
uruguay_plot.set_xlabel('Year',fontsize=14)
plt.show()

#####################################################
## A plot of flu trends in Romania & Uruguay together
#####################################################

#Plot and store the flu trend for Romania
ax = df_romania_FluTrends.plot(legend ='left', x='Date', y = '# of cases of flu in Romania', 
                               c='r',figsize=(10, 6), grid=True)

#Plot the flu trend for Uruguay; ax=ax plots the chart into the previous
df_uruguay_FluTrends.plot(x='Date', y = '# of cases of flu in Uruguay' , 
                          figsize=(10, 6), grid=True, ax=ax)

#Labeling of the graph

ax.set_title('Google FluTrends surveillance data - Across all years',fontsize=15, fontweight='bold')
ax.set_ylabel('Frequency (cases of flu)',fontsize=14)
ax.set_xlabel('Year',fontsize=14)
plt.show()

Assignment D

D (2) i. The yearly minimum and maximum for each country

#################################################################
##Introduce Column that is the year taken from the column "Date" 
#################################################################

df_romania_FluTrends['year'] = pd.DatetimeIndex(df_romania_FluTrends['Date']).year
df_uruguay_FluTrends['year'] = pd.DatetimeIndex(df_uruguay_FluTrends['Date']).year

######################################
## Create lists of min and max values
######################################

romania_yearly_min=df_romania_FluTrends.groupby(['year'])['# of cases of flu in Romania'].min()
romania_yearly_max=df_romania_FluTrends.groupby(['year'])['# of cases of flu in Romania'].max()

uruguay_yearly_min=df_uruguay_FluTrends.groupby(['year'])['# of cases of flu in Uruguay'].min()
uruguay_yearly_max=df_uruguay_FluTrends.groupby(['year'])['# of cases of flu in Uruguay'].max()

######################################
## Combine lists into one DataFrame
######################################

df_Rom_Uru_min_max_temp = pd.DataFrame([uruguay_yearly_min, romania_yearly_min,
                                     uruguay_yearly_max,romania_yearly_max],
####################
## Re-name indices                                    
####################
                                    
index=['Min. annual cases of flu (Uruguay min.)', 'Min. annual cases of flu (Romania min.)', 
       'Max. annual cases of flu (Uruguay max.)', 'Max. annual cases of flu (Romania max.)'])
#######################
## Transpose & format
######################

df_Rom_Uru_min_max_temp2=pd.DataFrame(df_Rom_Uru_min_max_temp.transpose())

df_Rom_Uru_min_max = pd.DataFrame(df_Rom_Uru_min_max_temp2, dtype='float')
pd.options.display.float_format = '{:2,.0f}'.format

df_Rom_Uru_min_max

Rom_Uru_min_max_temp_array=df_Rom_Uru_min_max_temp.values # convert dataframe into pandasnumarray

uru_min=Rom_Uru_min_max_temp_array[0,:]
rom_min=Rom_Uru_min_max_temp_array[1,:]
uru_max=Rom_Uru_min_max_temp_array[2,:]
rom_max=Rom_Uru_min_max_temp_array[3,:]

x = np.arange(2003.0, 2016.0, 1.0) # This is the x-axis scale

#########################
## These are the y-values
#########################

y1=uru_min
y2=rom_min
y3=uru_max
y4=rom_max

s = 100 # s = Size of plot points

##############################################
## Plot for each countries' min and max values
##############################################

plt.figure(figsize=(14, 7))

plt.scatter(x,y4, s, c='red', alpha=0.8, marker='^',label="Romania Max")
plt.scatter(x,y3, s, c="darkblue", alpha=0.8, marker='^',label="Uruguay Max")
plt.scatter(x,y2, s, c="red", alpha=0.8, marker='_',label="Romania Min")
plt.scatter(x,y1, s, c="darkblue", alpha=0.8, marker='_',label="Uruguay Min")

plt.title('Plot of yearly min. and max. number of flu cases in Romania & Uruguay', 
          fontsize=15, fontweight='bold')
plt.xlabel("Year",fontsize=14) # Label the x-axis
plt.ylabel("Number of flu cases",fontsize=14)  # Label the y-axis

plt.legend(loc='best') #Location of legend ensures that legend doesn't cover the data

plt.grid(True) #Include plotlines


plt.show()

Assignment D

D (2) ii. A mathematical function to approximate the seasonal trend of each country

romania_date_list= [i%52 for i in range(0, len(df_romania_FluTrends['Date']))]
romania_flu_nbrs = df_romania_FluTrends['# of cases of flu in Romania'].tolist()


# Fit the data to obtain the fitted coefficients [a,b,c,d,e]
# for the quartic poly a + bx^1 + cx^2 + dx^3 + ex^4 

x_r = romania_date_list
y_r = romania_flu_nbrs
deg = 4

romania_coef = poly.polyfit(x_r, y_r, deg, rcond=None, full=False, w=None)

reverse_romania_coef = romania_coef[::-1] # need to reverse order of coeff so you can use poly1d
eqn_romania = np.poly1d(reverse_romania_coef) # This is the suitable equation

###################################
# Print a list of the coefficients
###################################

print("""On fitting the Romania FluTrends data the fitted coefficients are given by:

%s""" % romania_coef)
print("""
""")

##################################
# Print the mathematical function
#################################

print("""A suitable mathematical function that can be used as an approximation for the 
seasonal trend of flu cases in Romania is given by the quartic polynomial:""")
print("""
""")
print("""%s""" % eqn_romania)

On fitting the Romania FluTrends data the fitted coefficients are given by:

[ 8.98348980e+02 -6.57942241e-01 -2.38970454e+00  1.00497858e-01
 -1.06055332e-03]


A suitable mathematical function that can be used as an approximation for the 
seasonal trend of flu cases in Romania is given by the quartic polynomial:


           4          3        2
-0.001061 x + 0.1005 x - 2.39 x - 0.6579 x + 898.3

# Evaluate the polynomial at points x=date_list 
# using the fitted coefficients c=coefficients

c_r=romania_coef
x_r=romania_date_list

romania_curve_values = poly.polyval(x_r, c_r, tensor=True)

df_romania_curve_values=pd.DataFrame(romania_curve_values,columns=['Romania seasonal trend approx'])

df_romania_no_date=pd.DataFrame(romania_flu_nbrs,columns=['Romania FluTrends data'])

#####################################################
## A plot of flu trends in Romania & Uruguay together
#####################################################

#Plot and store the flu trend for Romania
ax = df_romania_no_date.plot(c='r', alpha=0.8, figsize=(10, 6), grid=True)

#Plot the flu trend for fitted Romania curve; ax=ax plots the chart into the previous
df_romania_curve_values.plot(c='green', figsize=(10, 6), grid=True, ax=ax)

#Labeling of the graph

ax.set_title('Romania FluTrends Data: An approximation to the seasonal trend component',fontsize=15, fontweight='bold')
ax.set_ylabel('Frequency (cases of flu)',fontsize=14)

plt.xticks([]) #Remove x-ticks for presentation only

plt.show()

uruguay_date_list= [i%52 for i in range(0, len(df_uruguay_FluTrends['Date']))]
uruguay_flu_nbrs = df_uruguay_FluTrends['# of cases of flu in Uruguay'].tolist()


# Fit the data to obtain the fitted coefficients [a,b,c,d,e]
# for the quartic poly a + bx^1 + cx^2 + dx^3 + ex^4 

x_r = uruguay_date_list
y_r = uruguay_flu_nbrs
deg = 4

uruguay_coef = poly.polyfit(x_r, y_r, deg, rcond=None, full=False, w=None)

reverse_uruguay_coef = uruguay_coef[::-1] # need to reverse order of coeff so you can use poly1d
eqn_uruguay = np.poly1d(reverse_uruguay_coef) # This is the suitable equation

###################################
# Print a list of the coefficients
###################################

print("""On fitting the Uruguay FluTrends data the fitted coefficients are given by:

%s""" % uruguay_coef)
print("""
""")

##################################
# Print the mathematical function
#################################

print("""A suitable mathematical function that can be used as an approximation for the 
seasonal trend of flu cases in Uruguay is given by the quartic polynomial:""")
print("""
""")
print("""%s""" % eqn_uruguay)

On fitting the Uruguay FluTrends data the fitted coefficients are given by:

[ 6.78230464e+01  6.09321302e+00  2.24730822e-01 -1.87909316e-02
  2.38878639e-04]


A suitable mathematical function that can be used as an approximation for the 
seasonal trend of flu cases in Uruguay is given by the quartic polynomial:


           4           3          2
0.0002389 x - 0.01879 x + 0.2247 x + 6.093 x + 67.82

# Evaluate the polynomial at points x=date_list 
# using the fitted coefficients c=coefficients

c_r=uruguay_coef
x_r=uruguay_date_list

uruguay_curve_values = poly.polyval(x_r, c_r, tensor=True)

df_uruguay_curve_values=pd.DataFrame(uruguay_curve_values,columns=['Uruguay seasonal trend approx'])

df_uruguay_no_date=pd.DataFrame(uruguay_flu_nbrs,columns=['Uruguay FluTrends data'])

#####################################################
## A plot of flu trends in Uruguay & Uruguay together
#####################################################

#Plot and store the flu trend for Uruguay
ax = df_uruguay_no_date.plot(alpha=0.8, figsize=(10, 6), grid=True)

#Plot the flu trend for fitted Uruguay curve; ax=ax plots the chart into the previous
df_uruguay_curve_values.plot(c='orange', figsize=(10, 6), grid=True, ax=ax)

#Labeling of the graph

ax.set_title('Uruguay FluTrends Data: An approximation to the seasonal trend component',fontsize=15, fontweight='bold')
ax.set_ylabel('Frequency (cases of flu)',fontsize=14)

plt.xticks([]) #Remove x-ticks for presentation only

plt.show()

	PRACTICE	TOTAL NUMBER OF PRESCRIPTIONS	TOTAL COST OF PRESCRIPTIONS (£)
0	P82001	18,419	164,398
1	P82002	12,734	108,972
2	P82003	22,613	208,776
3	P82004	15,775	120,830
4	P82005	9,674	92,823
5	P82006	13,373	114,165
6	P82007	24,377	203,194
7	P82008	29,330	217,671
8	P82009	14,147	102,178
9	P82010	12,816	94,423
10	P82011	9,111	86,341
11	P82012	10,905	96,096
12	P82013	12,536	84,234
13	P82014	15,511	129,462
14	P82015	33,587	318,124
15	P82016	23,486	207,838
16	P82018	9,081	101,715
17	P82020	6,896	66,429
18	P82021	8,710	82,933
19	P82022	8,897	61,491
20	P82023	12,220	131,742
21	P82025	6,083	55,772
22	P82029	6,615	49,690
23	P82030	6,791	37,866
24	P82031	16,231	161,002
25	P82033	3,863	28,438
26	P82034	4,609	30,822
27	P82036	6,191	54,560
28	P82037	8,321	63,717
29	P82607	9,392	61,743
30	P82609	14,020	76,631
31	P82613	10,541	71,543
32	P82616	4,858	34,613
33	P82624	6,108	47,270
34	P82625	2,618	25,388
35	P82626	8,545	60,012
36	P82627	7,488	66,082
37	P82629	3,278	23,273
38	P82633	5,360	40,076
39	P82634	5,443	56,109
40	P82640	6,785	55,354
41	P82643	6,307	72,824
42	P82652	6,607	50,460
43	P82654	14	265
44	P82660	6,620	35,343
45	P82661	470	2,823
46	Y00186	1,547	11,418
47	Y00208	9	128
48	Y00215	70	146,873
49	Y00233	1	27
50	Y00448	150	3,483
51	Y00552	1	1
52	Y00747	165	7,256
53	Y02319	5,095	35,864
54	Y02790	4,155	29,414
55	Y02943	56	1,436
56	Y03079	22,209	211,644
57	Y03364	5,619	44,084
58	Y03366	5,030	39,415
59	Y03641	2,267	27,141
60	Y04078	2,921	63,437
61	Y04600	3,042	16,041

	BNF CODE	BNF NAME	TOTAL ITEMS DISPENSED	TOTAL ACTUAL COST (£)
Most prescribed	0103050P0AAAAAA	Omeprazole_Cap E/C 20mg	16,667	25,256
2nd most prescribed	0212000Y0AAADAD	Simvastatin_Tab 40mg	11,114	14,438
3rd	0407010H0AAAMAM	Paracet_Tab 500mg	9,999	32,997
4th	0209000A0AAABAB	Aspirin Disper_Tab 75mg	9,380	8,505
5th	0301011R0AAAPAP	Salbutamol_Inha 100mcg (200 D) CFF	7,154	12,298
6th	0206020A0AAAAAA	Amlodipine_Tab 5mg	6,718	7,549
7th	0601022B0AAABAB	Metformin HCl_Tab 500mg	5,824	25,465
8th	0202010B0AAABAB	Bendroflumethiazide_Tab 2.5mg	5,789	6,529
9th	0103050L0AAAAAA	Lansoprazole_Cap 30mg (E/C Gran)	5,677	10,022
10th	0205051R0AAADAD	Ramipril_Cap 10mg	5,551	9,133

	BNF CODE	BNF NAME	TOTAL ITEMS DISPENSED	TOTAL ACTUAL COST (£)
Least prescribed	0407010Q0AAAAAA	Co-Proxamol_Tab 32.5mg/325mg	1	0.18
2nd least prescribed	20100000551	Unspecified Surg Adh Tape 1.25cm x 5m Pe	1	0.26
3rd	091000000BBPPA0	Valupak_Multivit & Mineral Tab OAD	1	0.32
4th	20090000936	CliniFast 5cm x 1m (Green) Stkntte Elasc	1	0.55
5th	20090000451	Unspecified Stkntte 6.25cm x 0.5m Elasct	1	0.55
6th	0906011H0AAAGAG	Cod Liver Oil_Cap 1g	1	0.57
7th	0106020M0AAAEAE	Senna_Oral Soln 7.5mg/5ml	1	0.61
8th	20040200300	Ribbon Gz 1.25cm x 5m Ster Absorbent Cot	1	0.80
9th	20090000805	Coverflex 5cm x 1m (Green) Stkntte Elasc	1	0.80
10th	0212000Y0BDABAB	Simvador_Tab 20mg	1	0.81

	PRACTICE	Total Monthly Spend on all Prescriptions (£)	Most prescribed	2nd most prescribed	3rd	4th	5th	6th	7th	8th	9th	10th
0	P82001	164,398	15.4%	8.8%	20.1%	5.2%	7.5%	4.6%	15.5%	4.0%	6.1%	5.6%
1	P82002	108,972	23.2%	13.2%	30.3%	7.8%	11.3%	6.9%	23.4%	6.0%	9.2%	8.4%
2	P82003	208,776	12.1%	6.9%	15.8%	4.1%	5.9%	3.6%	12.2%	3.1%	4.8%	4.4%
3	P82004	120,830	20.9%	11.9%	27.3%	7.0%	10.2%	6.2%	21.1%	5.4%	8.3%	7.6%
4	P82005	92,823	27.2%	15.6%	35.5%	9.2%	13.2%	8.1%	27.4%	7.0%	10.8%	9.8%
5	P82006	114,165	22.1%	12.6%	28.9%	7.4%	10.8%	6.6%	22.3%	5.7%	8.8%	8.0%
6	P82007	203,194	12.4%	7.1%	16.2%	4.2%	6.1%	3.7%	12.5%	3.2%	4.9%	4.5%
7	P82008	217,671	11.6%	6.6%	15.2%	3.9%	5.6%	3.5%	11.7%	3.0%	4.6%	4.2%
8	P82009	102,178	24.7%	14.1%	32.3%	8.3%	12.0%	7.4%	24.9%	6.4%	9.8%	8.9%
9	P82010	94,423	26.7%	15.3%	34.9%	9.0%	13.0%	8.0%	27.0%	6.9%	10.6%	9.7%
10	P82011	86,341	29.3%	16.7%	38.2%	9.9%	14.2%	8.7%	29.5%	7.6%	11.6%	10.6%
11	P82012	96,096	26.3%	15.0%	34.3%	8.9%	12.8%	7.9%	26.5%	6.8%	10.4%	9.5%
12	P82013	84,234	30.0%	17.1%	39.2%	10.1%	14.6%	9.0%	30.2%	7.8%	11.9%	10.8%
13	P82014	129,462	19.5%	11.2%	25.5%	6.6%	9.5%	5.8%	19.7%	5.0%	7.7%	7.1%
14	P82015	318,124	7.9%	4.5%	10.4%	2.7%	3.9%	2.4%	8.0%	2.1%	3.2%	2.9%
15	P82016	207,838	12.2%	6.9%	15.9%	4.1%	5.9%	3.6%	12.3%	3.1%	4.8%	4.4%
16	P82018	101,715	24.8%	14.2%	32.4%	8.4%	12.1%	7.4%	25.0%	6.4%	9.9%	9.0%
17	P82020	66,429	38.0%	21.7%	49.7%	12.8%	18.5%	11.4%	38.3%	9.8%	15.1%	13.7%
18	P82021	82,933	30.5%	17.4%	39.8%	10.3%	14.8%	9.1%	30.7%	7.9%	12.1%	11.0%
19	P82022	61,491	41.1%	23.5%	53.7%	13.8%	20.0%	12.3%	41.4%	10.6%	16.3%	14.9%
20	P82023	131,742	19.2%	11.0%	25.0%	6.5%	9.3%	5.7%	19.3%	5.0%	7.6%	6.9%
21	P82025	55,772	45.3%	25.9%	59.2%	15.2%	22.1%	13.5%	45.7%	11.7%	18.0%	16.4%
22	P82029	49,690	50.8%	29.1%	66.4%	17.1%	24.7%	15.2%	51.2%	13.1%	20.2%	18.4%
23	P82030	37,866	66.7%	38.1%	87.1%	22.5%	32.5%	19.9%	67.3%	17.2%	26.5%	24.1%
24	P82031	161,002	15.7%	9.0%	20.5%	5.3%	7.6%	4.7%	15.8%	4.1%	6.2%	5.7%
25	P82033	28,438	88.8%	50.8%	116.0%	29.9%	43.2%	26.5%	89.5%	23.0%	35.2%	32.1%
26	P82034	30,822	81.9%	46.8%	107.1%	27.6%	39.9%	24.5%	82.6%	21.2%	32.5%	29.6%
27	P82036	54,560	46.3%	26.5%	60.5%	15.6%	22.5%	13.8%	46.7%	12.0%	18.4%	16.7%
28	P82037	63,717	39.6%	22.7%	51.8%	13.3%	19.3%	11.8%	40.0%	10.2%	15.7%	14.3%
29	P82607	61,743	40.9%	23.4%	53.4%	13.8%	19.9%	12.2%	41.2%	10.6%	16.2%	14.8%
30	P82609	76,631	33.0%	18.8%	43.1%	11.1%	16.0%	9.9%	33.2%	8.5%	13.1%	11.9%
31	P82613	71,543	35.3%	20.2%	46.1%	11.9%	17.2%	10.6%	35.6%	9.1%	14.0%	12.8%
32	P82616	34,613	73.0%	41.7%	95.3%	24.6%	35.5%	21.8%	73.6%	18.9%	29.0%	26.4%
33	P82624	47,270	53.4%	30.5%	69.8%	18.0%	26.0%	16.0%	53.9%	13.8%	21.2%	19.3%
34	P82625	25,388	99.5%	56.9%	130.0%	33.5%	48.4%	29.7%	100.3%	25.7%	39.5%	36.0%
35	P82626	60,012	42.1%	24.1%	55.0%	14.2%	20.5%	12.6%	42.4%	10.9%	16.7%	15.2%
36	P82627	66,082	38.2%	21.8%	49.9%	12.9%	18.6%	11.4%	38.5%	9.9%	15.2%	13.8%
37	P82629	23,273	108.5%	62.0%	141.8%	36.5%	52.8%	32.4%	109.4%	28.1%	43.1%	39.2%
38	P82633	40,076	63.0%	36.0%	82.3%	21.2%	30.7%	18.8%	63.5%	16.3%	25.0%	22.8%
39	P82634	56,109	45.0%	25.7%	58.8%	15.2%	21.9%	13.5%	45.4%	11.6%	17.9%	16.3%
40	P82640	55,354	45.6%	26.1%	59.6%	15.4%	22.2%	13.6%	46.0%	11.8%	18.1%	16.5%
41	P82643	72,824	34.7%	19.8%	45.3%	11.7%	16.9%	10.4%	35.0%	9.0%	13.8%	12.5%
42	P82652	50,460	50.1%	28.6%	65.4%	16.9%	24.4%	15.0%	50.5%	12.9%	19.9%	18.1%
43	P82654	265	9,526.7%	5,446.0%	12,446.5%	3,208.0%	4,638.9%	2,847.6%	9,605.5%	2,462.9%	3,780.2%	3,445.0%
44	P82660	35,343	71.5%	40.9%	93.4%	24.1%	34.8%	21.4%	72.1%	18.5%	28.4%	25.8%
45	P82661	2,823	894.7%	511.5%	1,168.9%	301.3%	435.7%	267.4%	902.1%	231.3%	355.0%	323.5%
46	Y00186	11,418	221.2%	126.4%	289.0%	74.5%	107.7%	66.1%	223.0%	57.2%	87.8%	80.0%
47	Y00208	128	19,808.9%	11,323.8%	25,879.9%	6,670.3%	9,645.6%	5,921.0%	19,972.6%	5,121.1%	7,860.2%	7,163.1%
48	Y00215	146,873	17.2%	9.8%	22.5%	5.8%	8.4%	5.1%	17.3%	4.4%	6.8%	6.2%
49	Y00233	27	91,941.4%	52,558.4%	120,119.7%	30,959.6%	44,769.5%	27,481.9%	92,701.2%	23,769.4%	36,482.5%	33,246.9%
50	Y00448	3,483	725.2%	414.5%	947.4%	244.2%	353.1%	216.8%	731.2%	187.5%	287.8%	262.2%
51	Y00552	1	2,405,361.0%	1,375,028.6%	3,142,559.0%	809,961.9%	1,171,256.2%	718,979.0%	2,425,241.0%	621,853.3%	954,450.5%	869,801.0%
52	Y00747	7,256	348.1%	199.0%	454.7%	117.2%	169.5%	104.0%	350.9%	90.0%	138.1%	125.9%
53	Y02319	35,864	70.4%	40.3%	92.0%	23.7%	34.3%	21.0%	71.0%	18.2%	27.9%	25.5%
54	Y02790	29,414	85.9%	49.1%	112.2%	28.9%	41.8%	25.7%	86.6%	22.2%	34.1%	31.0%
55	Y02943	1,436	1,758.8%	1,005.4%	2,297.9%	592.3%	856.4%	525.7%	1,773.4%	454.7%	697.9%	636.0%
56	Y03079	211,644	11.9%	6.8%	15.6%	4.0%	5.8%	3.6%	12.0%	3.1%	4.7%	4.3%
57	Y03364	44,084	57.3%	32.8%	74.8%	19.3%	27.9%	17.1%	57.8%	14.8%	22.7%	20.7%
58	Y03366	39,415	64.1%	36.6%	83.7%	21.6%	31.2%	19.2%	64.6%	16.6%	25.4%	23.2%
59	Y03641	27,141	93.1%	53.2%	121.6%	31.3%	45.3%	27.8%	93.8%	24.1%	36.9%	33.6%
60	Y04078	63,437	39.8%	22.8%	52.0%	13.4%	19.4%	11.9%	40.1%	10.3%	15.8%	14.4%
61	Y04600	16,041	157.4%	90.0%	205.7%	53.0%	76.7%	47.1%	158.7%	40.7%	62.5%	56.9%

	Amount Spent for the Month (£)	as a percentage of Total Bolton Prescription Spend
All medications prescribed in Bolton	4,693,171	100.00%
Top 10 most frequently prescribed medications	152,192	3.24%
Most prescribed	25,256	0.54%
2nd most prescribed	14,438	0.31%
3rd	32,997	0.70%
4th	8,505	0.18%
5th	12,298	0.26%
6th	7,549	0.16%
7th	25,465	0.54%
8th	6,529	0.14%
9th	10,022	0.21%
10th	9,133	0.19%

	Total Amount Spent across all Bolton (£)
10 most frequently prescribed medications	152,192
10 least frequently prescribed medications	5

	BNF CODE	BNF NAME	ACT COST PER ITEM (£)
Most expensive	1001030D0AAACAC	Etanercept_Inj 50mg/ml Pfs	3,967
2nd most expensive	1001030S0AAABAB	Adalimumab_Inj 40mg/0.8ml Pf Pen	3,456
3rd	1001030Y0AAAAAA	Certolizumab Pegol_Inj 200mg/1ml Pfs	3,085
4th	1001030D0AAAEAE	Etanercept_Inj 50mg/ml 1ml Pf Pen	2,777
5th	1001030D0BBAFAE	Enbrel_Inj 50mg/ml 1ml Pf MyClic Pen	2,645
6th	1001030W0AAADAD	Tocilizumab_Inj 162mg/0.9ml Pfs	2,533
7th	0408010N0AACLCL	Phenobarb_Liq Spec 50mg/5ml	1,553
8th	0601011R0BBABAB	Ins Humulin R_500u/ml 20ml Vl	1,493
9th	0604011D0AAACAC	Ethinylestr_Tab 50mcg	1,479
10th	0803043N0BBAGAG	Sandostatin LAR_Inj 20mg Vl + Dil	1,478

	CITIES	NUMBER_OF_GP_PRACTICES
0	BIRMINGHAM	331
1	MANCHESTER	187
2	NOTTINGHAM	153
3	LEEDS	143
4	LIVERPOOL	129
5	BRISTOL	124
6	LEICESTER	117
7	COVENTRY	103
8	BRADFORD	84
9	PLYMOUTH	67
10	BOLTON	62
11	WOLVERHAMPTON	62
12	DONCASTER	55
13	SHEFFIELD	53
14	BRIGHTON	49
15	SOUTHAMPTON	44
16	LUTON	43
17	CAMBRIDGE	40
18	PORTSMOUTH	34
19	BOURNEMOUTH	28

	count	mean	std	min	25%	50%	75%	max
National Level	7,701.00	21.32	665.57	0.00	10.69	13.21	15.50	58,319.78
Bolton GP Practices	50.00	14.27	2.83	7.78	12.54	14.72	15.58	22.11

	count	mean	std	min	25%	50%	75%	max
National Level	9,906	76,989	69,967	0	21,719	61,983	117,817	923,786
Bolton GP Practices	62	75,696	65,599	1	31,770	60,752	100,311	318,124

	PRACTICE	PRACTICE NAME	POSTCODE	IMD RANK	IMD DECILE
0	P82001	THE DUNSTAN PARTNERSHIP	BL2 6NT	3,900	2
1	P82002	PIKES LANE 1	BL3 5HP	4,389	2
2	P82003	KILDONAN HOUSE	BL6 5NW	18,170	6
3	P82004	SWAN LANE MEDICAL CENTRE	BL3 6TL	7,372	3
4	P82005	STABLE FOLD SURGERY	BL5 3SF	17,059	6
5	P82006	DR MALHOTRA & PARTNERS	BL6 7AS	15,089	5
6	P82007	KEARSLEY MEDICAL CENTRE	BL4 8EP	8,370	3
7	P82008	STONEHILL MEDICAL CENTRE	BL4 9QZ	6,966	3
8	P82009	ST HELENS ROAD PRACTICE	BL3 3RR	13,656	5
9	P82010	DALEFIELD SURGERY	BL1 4JP	1,294	1
10	P82011	TONGE FOLD HEALTH CENTRE	BL2 6DY	7,955	3
11	P82012	LEVER CHAMBERS 1	BL1 1SQ	2,131	1
12	P82013	LEVER CHAMBERS 2	BL3 6RN	1,238	1
13	P82014	SPRING HOUSE SURGERY	BL1 6AF	27,107	9
14	P82015	UNSWORTH GROUP PRACTICE	BL5 3UB	30,911	10
15	P82016	HARWOOD MEDICAL CENTRE	BL2 3HQ	23,271	8
16	P82018	THE ALASTAIR ROSS MEDICAL PRACTICE	BL2 6NT	3,900	2
17	P82020	LITTLE LEVER HEALTH CENTRE 1	BL3 1JF	23,116	8
18	P82021	KIRBY-CROMPTON HEALTH CENTRE	BL1 8UP	6,766	3
19	P82022	HALLIWELL SURGERY 1	BL1 3RG	985	1
20	P82023	MANDALAY MEDICAL CENTRE	BL1 7LR	29,855	10
21	P82025	BURNSIDE SURGERY	BL1 8TU	4,386	2
22	P82029	HALLIWELL SURGERY 2	BL1 3RG	985	1
23	P82030	DEANE MEDICAL CENTRE	BL3 3QB	3,711	2
24	P82031	HEATON MEDICAL CENTRE	BL1 5PU	24,314	8
25	P82033	BRADFORD STREET SURGERY	BL2 1HT	2,245	1
26	P82034	EDGWORTH MEDICAL CENTRE	BL7 9RG	16,159	5
27	P82036	LITTLE LEVER HEALTH CENTRE 2	BL3 1JF	23,116	8
28	P82037	FARNWORTH HEALTH CENTRE 1	BL4 9AL	5,200	2
29	P82607	WALMSLEY-CROMPTON HEALTH CENTRE	BL1 8UP	6,766	3
30	P82609	SHANTI MEDICAL CENTRE	BL3 3PH	3,365	2
31	P82613	SPRING VIEW MEDICAL CENTRE	BL3 1HQ	5,598	2
32	P82616	BEEHIVE SURGERY	BL3 2JR	1,238	1
33	P82624	ORIENT HOUSE MEDICAL CENTRE	BL3 5QE	3,636	2
34	P82625	CHARLOTTE STREET SURGERY	BL1 8TU	4,386	2
35	P82626	HALLIWELL SURGERY 3	BL1 3RG	985	1
36	P82627	CORNERSTONE SURGERY	BL1 6AH	5,411	2
37	P82629	PIKES LANE 2	BL3 5HP	4,389	2
38	P82633	GREAT LEVER HEALTH CENTRE 1	BL3 6RN	1,238	1
39	P82634	WYRESDALE ROAD SURGERY	BL1 4DN	3,175	1
40	P82640	AL FAL MEDICAL GROUP	BL3 5HP	4,389	2
41	P82643	EGERTON/DUNSCAR HEALTH CENTRE	BL7 9RG	16,159	5
42	P82652	FARNWORTH HEALTH CENTRE 2	BL4 9AH	1,365	1
43	P82654	BOLTON HOSPICE	BL1 4QT	2,131	1
44	P82660	DEANE CLINIC 1	BL3 4LU	3,636	2
45	P82661	INTERMEDIATE CARE	BL1 3EJ	6,819	3
46	Y00186	3D MEDICAL CENTRE	BL3 5DP	4,389	2
47	Y00208	TIER 2 DERMATOLOGY SERVICE	BL1 1SQ	2,131	1
48	Y00215	ORTHOPAEDIC & RHEUMATOLOGY	BL1 1SQ	2,131	1
49	Y00233	THE PARALLEL	BL1 1HU	488	1
50	Y00448	DIABETES CENTRE	BL1 4AL	201	1
51	Y00552	MINOR TREATMENT CLINIC	BL1 1SQ	2,131	1
52	Y00747	HALLIWELL HEALTH & CHILDREN'S CENTRE	BL1 3SQ	985	1
53	Y02319	BOLTON GENERAL PRACTICE	BL1 2AY	2,131	1
54	Y02790	BOLTON MEDICAL CENTRE	BL3 6PY	1,238	1
55	Y02943	NEUROLOGY LONG TERM CONDITIONS	BL2 6NT	3,900	2
56	Y03079	BOLTON COMMUNITY PRACTICE	BL1 8TU	4,386	2
57	Y03364	GREAT LEVER PRACTICE	BL3 6RN	1,238	1
58	Y03366	OLIVE FAMILY PRACTICE	BL3 5HP	4,389	2
59	Y03641	BOLTON COMMUNITY DRUG AND ALCOHOL SERV	BL1 4TQ	2,131	1
60	Y04078	NON MEDICAL PRESCRIBING	BL4 0JR	20,709	7
61	Y04600	BARDOC GP OOH	BL1 8TT	985	1

	PRACTICE	TOTAL NUMBER OF PATIENTS	TOTAL COST OF STATIN PRESCRIPTIONS (£)	STATIN COST PER PATIENT (£)	Decile
0	P81633	3,083	1,052	0.34	9
1	L83145	2,775	947	0.34	9
2	G81004	10,539	3,597	0.34	9
3	P84012	10,654	3,637	0.34	9
4	L84026	10,165	3,470	0.34	9
5	D82629	3,503	1,196	0.34	9
6	A88015	7,352	2,510	0.34	9
7	P82031	10,423	3,560	0.34	9
8	P81073	10,929	3,733	0.34	9
9	C83649	4,330	1,479	0.34	9
10	F81087	13,762	4,704	0.34	9
11	B81019	2,709	926	0.34	9
12	P86014	8,497	2,906	0.34	9
13	B82037	8,900	3,045	0.34	9
14	A85605	2,312	791	0.34	9
15	C83641	4,185	1,432	0.34	9
16	L81046	10,242	3,508	0.34	9
17	N82657	1,136	389	0.34	9
18	G81048	7,414	2,541	0.34	9
19	C88072	12,415	4,256	0.34	9
20	M87016	6,166	2,114	0.34	9
21	B81689	2,010	689	0.34	9
22	H82034	13,589	4,665	0.34	9
23	G81077	6,949	2,387	0.34	9
24	P85601	2,196	754	0.34	9
25	G82687	4,162	1,430	0.34	9
26	N84025	8,672	2,981	0.34	9
27	Y02511	710	244	0.34	9
28	L85044	5,578	1,918	0.34	9
29	G82044	7,026	2,416	0.34	9
30	L83055	10,924	3,758	0.34	9
31	L84042	5,123	1,763	0.34	9
32	B87033	3,768	1,297	0.34	9
33	A83019	5,161	1,776	0.34	9
34	C82025	10,412	3,584	0.34	9
35	P84045	5,090	1,752	0.34	9
36	G82037	15,989	5,505	0.34	9
37	D83017	12,454	4,291	0.34	9
38	F82668	2,754	949	0.34	9
39	A89614	4,021	1,385	0.34	9
40	P92023	7,951	2,740	0.34	9
41	C85004	16,608	5,734	0.35	9
42	D82028	5,551	1,917	0.35	9
43	P92001	5,470	1,889	0.35	9
44	A89011	4,016	1,387	0.35	9
45	H82069	7,165	2,475	0.35	9
46	G82158	5,000	1,727	0.35	9
47	B83056	7,390	2,556	0.35	9
48	G82212	6,311	2,183	0.35	9
49	F86083	6,913	2,392	0.35	9
50	F85632	1,801	623	0.35	9
51	H82086	2,392	828	0.35	9
52	B83029	9,280	3,214	0.35	9
53	C83637	7,257	2,514	0.35	9
54	C86009	8,080	2,799	0.35	9
55	G82165	9,094	3,152	0.35	9
56	H85086	1,292	448	0.35	9
57	N83019	10,288	3,567	0.35	9
58	C83005	7,329	2,542	0.35	9
59	N83060	3,889	1,349	0.35	9
60	C88656	3,054	1,060	0.35	9
61	F81696	1,978	686	0.35	9
62	A81008	4,027	1,398	0.35	9
63	M82048	5,219	1,812	0.35	9
64	P82009	6,014	2,088	0.35	9
65	L85009	10,068	3,496	0.35	9
66	G81685	1,540	535	0.35	9
67	G82038	9,658	3,358	0.35	9
68	H85056	1,517	527	0.35	9
69	B86001	2,405	836	0.35	9
70	F81142	6,165	2,143	0.35	9
71	E86028	7,639	2,657	0.35	9
72	P86608	4,259	1,483	0.35	9
73	C83043	10,163	3,540	0.35	9
74	D82025	9,046	3,151	0.35	9
75	B85001	7,529	2,623	0.35	9
76	L84616	5,843	2,037	0.35	9
77	C84688	9,650	3,366	0.35	9
78	N84621	2,557	892	0.35	9
79	J82083	8,488	2,962	0.35	9
80	P92028	4,389	1,532	0.35	9
81	H82031	5,449	1,903	0.35	9
82	M86007	6,309	2,204	0.35	9
83	G82799	2,565	896	0.35	9
84	M82022	12,877	4,499	0.35	9
85	H82035	11,154	3,899	0.35	9
86	P88021	7,380	2,580	0.35	9
87	B81602	2,486	869	0.35	9
88	B82032	10,225	3,577	0.35	9
89	B87007	11,572	4,049	0.35	9
90	N83024	7,390	2,586	0.35	9
91	P92034	5,886	2,061	0.35	9
92	N82109	3,070	1,075	0.35	9
93	E87026	2,860	1,002	0.35	9
94	B82073	9,670	3,389	0.35	9
95	G81041	16,522	5,792	0.35	9
96	L84057	5,894	2,067	0.35	9
97	F81168	3,755	1,318	0.35	9
98	H85691	157	55	0.35	9
99	C82081	2,426	853	0.35	9
100	C87621	1,787	629	0.35	9
101	Y03222	5,565	1,959	0.35	9
102	A81014	3,821	1,345	0.35	9
103	C83020	8,814	3,103	0.35	9
104	M85176	4,065	1,431	0.35	9
105	P81692	1,601	565	0.35	9
106	M83146	7,963	2,808	0.35	9
107	A84609	1,618	571	0.35	9
108	B82033	10,515	3,709	0.35	9
109	A83021	13,867	4,892	0.35	9
110	P85016	14,269	5,036	0.35	9
111	M83047	4,718	1,666	0.35	9
112	P91009	11,809	4,169	0.35	9
113	P82001	10,434	3,683	0.35	9
114	G81026	7,296	2,576	0.35	9
115	B87005	10,843	3,830	0.35	9
116	B82010	7,055	2,492	0.35	9
117	P81138	2,769	979	0.35	9
118	F82648	2,704	956	0.35	9
119	L83070	4,857	1,718	0.35	9
120	N81626	3,899	1,379	0.35	9
121	M83110	2,183	772	0.35	9
122	P91615	5,511	1,953	0.35	9
123	C83052	3,385	1,200	0.35	9
124	C84008	4,849	1,719	0.35	9
125	N85052	2,539	902	0.36	9
126	N82082	4,707	1,673	0.36	9
127	M89009	10,267	3,650	0.36	9
128	P92004	8,039	2,858	0.36	9
129	G83064	2,314	824	0.36	9
130	H82032	12,314	4,386	0.36	9
131	A83007	9,626	3,431	0.36	9
132	M87020	8,934	3,185	0.36	9
133	G82072	4,009	1,429	0.36	9
134	A89001	14,058	5,014	0.36	9
135	F81040	3,959	1,412	0.36	9
136	M87638	3,501	1,249	0.36	9
137	P85606	2,994	1,068	0.36	9
138	M82010	17,470	6,235	0.36	9
139	C85018	6,599	2,356	0.36	9
140	P86009	9,638	3,441	0.36	9
141	N85028	7,126	2,546	0.36	9
142	F82017	8,102	2,894	0.36	9
143	C82022	6,826	2,439	0.36	9
144	P85011	7,923	2,832	0.36	9
145	A84042	1,994	713	0.36	9
146	A82040	2,613	934	0.36	9
147	P85007	2,325	832	0.36	9
148	F81741	5,212	1,865	0.36	9
149	Y02461	3,273	1,173	0.36	9
150	A89008	4,934	1,769	0.36	9
151	P88018	8,388	3,008	0.36	9
152	G82197	2,610	936	0.36	9
153	P89013	5,482	1,966	0.36	9
154	G83007	5,326	1,910	0.36	9
155	L81030	6,619	2,376	0.36	9
156	P89003	9,776	3,514	0.36	9
157	F81613	6,913	2,485	0.36	9
158	L85056	5,767	2,073	0.36	9
159	A89018	4,548	1,636	0.36	9
160	K81077	10,786	3,881	0.36	9
161	D82046	4,177	1,504	0.36	9
162	E87017	2,510	905	0.36	9
163	P82023	9,527	3,434	0.36	9
164	Y04945	3,837	1,383	0.36	9
165	F81101	5,802	2,093	0.36	9
166	M87014	6,827	2,463	0.36	9
167	J84007	6,200	2,237	0.36	9
168	N82003	3,441	1,243	0.36	9
169	P84605	2,095	757	0.36	9
170	P92038	3,100	1,120	0.36	9
171	C82015	2,361	853	0.36	9
172	L85601	2,652	959	0.36	9
173	K81103	7,629	2,759	0.36	9
174	L84014	8,912	3,223	0.36	9
175	A83075	4,713	1,705	0.36	9
176	D83056	8,161	2,953	0.36	9
177	G82090	14,438	5,226	0.36	9
178	K81084	6,177	2,237	0.36	9
179	L84606	4,618	1,673	0.36	9
180	M92026	3,884	1,408	0.36	9
181	C86003	9,546	3,462	0.36	9
182	F82624	3,590	1,302	0.36	9
183	A89024	5,612	2,036	0.36	9
184	M87010	8,301	3,017	0.36	9
185	B87013	13,196	4,796	0.36	9
186	M85770	2,292	834	0.36	9
187	Y03364	1,994	726	0.36	9
188	B86625	3,753	1,366	0.36	9
189	J82156	7,142	2,600	0.36	9
190	C86001	10,695	3,893	0.36	9
191	M83079	8,860	3,226	0.36	9
192	M92044	4,265	1,554	0.36	9
193	P86016	6,967	2,540	0.36	9
194	P83009	7,669	2,799	0.36	9
195	J81069	4,660	1,702	0.37	9
196	D82015	8,667	3,166	0.37	9
197	B81030	12,656	4,624	0.37	9
198	M88032	7,573	2,768	0.37	9
199	B81066	2,298	840	0.37	9
200	P88044	4,701	1,719	0.37	9
201	L84023	13,093	4,788	0.37	9
202	L83089	11,871	4,341	0.37	9
203	M86638	5,068	1,854	0.37	9
204	L83094	7,928	2,901	0.37	9
205	M88633	3,769	1,381	0.37	9
206	M81094	8,222	3,017	0.37	9
207	C88090	4,468	1,640	0.37	9
208	P92012	4,861	1,785	0.37	9
209	M84062	4,705	1,730	0.37	9
210	M91019	14,081	5,177	0.37	9
211	B81004	12,399	4,566	0.37	9
212	A88012	5,891	2,169	0.37	9
213	P89006	11,263	4,148	0.37	9
214	G82123	4,961	1,827	0.37	9
215	B81065	11,728	4,320	0.37	9
216	B82079	3,223	1,188	0.37	9
217	Y03079	11,556	4,261	0.37	9
218	C87604	5,700	2,104	0.37	9
219	M81604	5,065	1,870	0.37	9
220	E85016	10,521	3,885	0.37	9
221	P88015	12,857	4,748	0.37	9
222	P81033	7,910	2,922	0.37	9
223	P92633	1,961	725	0.37	9
224	M81083	4,822	1,784	0.37	9
225	G82686	6,399	2,368	0.37	9
226	L83008	10,402	3,849	0.37	9
227	N83054	3,583	1,327	0.37	9
228	B81003	13,139	4,867	0.37	9
229	B85646	6,108	2,265	0.37	9
230	B82106	3,294	1,222	0.37	9
231	E85023	2,319	862	0.37	9
232	N82665	4,903	1,822	0.37	9
233	F81675	2,945	1,094	0.37	9
234	P81031	7,893	2,933	0.37	9
235	K81027	11,779	4,379	0.37	9
236	J82100	9,385	3,490	0.37	9
237	Y04883	2,451	911	0.37	9
238	P88610	3,351	1,248	0.37	9
239	A83051	10,208	3,801	0.37	9
240	P84041	6,441	2,400	0.37	9
241	C83063	15,637	5,827	0.37	9
242	P81738	1,177	439	0.37	9
243	C84033	7,638	2,850	0.37	9
244	H81611	1,909	713	0.37	9
245	Y05008	2,416	902	0.37	9
246	Y04884	3,878	1,448	0.37	9
247	G82696	2,200	822	0.37	9
248	B83022	9,081	3,393	0.37	9
249	C86621	4,819	1,802	0.37	9
250	B82064	3,370	1,265	0.38	9
251	L85007	11,045	4,146	0.38	9
252	A81608	2,955	1,109	0.38	9
253	M88004	15,226	5,716	0.38	9
254	F81026	7,513	2,821	0.38	9
255	L83624	5,771	2,168	0.38	9
256	L81119	5,045	1,896	0.38	9
257	M83041	5,427	2,040	0.38	9
258	Y02212	2,654	998	0.38	9
259	G82111	10,121	3,805	0.38	9
260	A89036	2,068	778	0.38	9
261	H82043	11,898	4,476	0.38	9
262	G83027	10,191	3,835	0.38	9
263	A88010	3,075	1,158	0.38	9
264	P85608	2,708	1,020	0.38	9
265	F82649	4,900	1,846	0.38	9
266	H82038	13,010	4,903	0.38	9
267	A83603	2,735	1,031	0.38	9
268	G81031	10,346	3,907	0.38	9
269	L83118	6,729	2,541	0.38	9
270	M83013	4,945	1,869	0.38	9
271	M88628	2,926	1,106	0.38	9
272	G82648	3,344	1,264	0.38	9
273	M83629	3,271	1,238	0.38	9
274	C84635	2,671	1,011	0.38	9
275	B87011	13,052	4,943	0.38	9
276	L82066	6,474	2,454	0.38	9
277	L83022	7,327	2,778	0.38	9
278	B86028	3,319	1,259	0.38	9
279	H82095	10,167	3,860	0.38	9
280	P82624	3,094	1,175	0.38	9
281	L83074	14,372	5,459	0.38	9
282	P86015	4,225	1,606	0.38	9
283	H82037	12,156	4,622	0.38	9
284	B81682	5,220	1,987	0.38	9
285	C83014	9,536	3,630	0.38	9
286	N84012	12,583	4,792	0.38	9
287	M83074	13,665	5,206	0.38	9
288	B85606	4,905	1,869	0.38	9
289	G84028	6,305	2,404	0.38	9
290	C83024	3,649	1,392	0.38	9
291	G81085	3,453	1,317	0.38	9
292	B81090	4,508	1,720	0.38	9
293	A83009	8,733	3,337	0.38	9
294	M83021	9,996	3,820	0.38	9
295	P88034	3,713	1,419	0.38	9
296	Y00182	36	14	0.38	9
297	L83125	2,983	1,140	0.38	9
298	M84620	2,556	978	0.38	9
299	M88620	2,210	845	0.38	9
300	K83069	2,337	894	0.38	9
301	A81066	4,281	1,639	0.38	9
302	L83657	4,129	1,582	0.38	9
303	D83054	3,876	1,485	0.38	9
304	B85610	8,802	3,373	0.38	9
305	C83071	7,288	2,793	0.38	9
306	F81126	9,297	3,571	0.38	9
307	F81177	2,826	1,086	0.38	9
308	P91012	6,277	2,413	0.38	9
309	C83038	8,169	3,142	0.38	9
310	G84033	6,924	2,664	0.38	9
311	D82049	3,506	1,350	0.39	9
312	E85035	4,562	1,757	0.39	9
313	M88010	8,514	3,280	0.39	9
314	G81084	5,095	1,964	0.39	9
315	L83030	8,173	3,152	0.39	9
316	L83064	11,928	4,601	0.39	9
317	C86039	5,518	2,129	0.39	9
318	F81147	5,087	1,963	0.39	9
319	C88074	5,078	1,960	0.39	9
320	D83019	7,258	2,802	0.39	9
321	M91640	3,399	1,312	0.39	9
322	F81018	7,557	2,919	0.39	9
323	B86675	3,571	1,380	0.39	9
324	A86035	2,625	1,016	0.39	9
325	P89008	7,197	2,785	0.39	9
326	M85142	1,463	566	0.39	9
327	Y05021	2,444	946	0.39	9
328	P89027	4,422	1,716	0.39	9
329	B81663	2,403	932	0.39	9
330	C83008	6,260	2,430	0.39	9
331	M81608	3,002	1,166	0.39	9
332	N82048	6,824	2,651	0.39	9
333	C85014	8,720	3,389	0.39	9
334	B81006	8,527	3,315	0.39	9
335	P81121	1,790	697	0.39	9
336	M88635	1,840	716	0.39	9
337	N81038	6,613	2,577	0.39	9
338	L83148	3,739	1,458	0.39	9
339	B83061	3,498	1,364	0.39	9
340	M88021	5,855	2,285	0.39	9
341	P88013	13,590	5,304	0.39	9
342	G81096	3,635	1,420	0.39	9
343	H81107	6,931	2,709	0.39	9
344	C83633	4,187	1,637	0.39	9
345	P88624	2,155	843	0.39	9
346	B83624	18,101	7,079	0.39	9
347	C83002	8,101	3,170	0.39	9
348	C86616	4,148	1,624	0.39	9
349	C83006	3,907	1,530	0.39	9
350	M84019	15,336	6,006	0.39	9
351	F86635	2,379	933	0.39	9
352	P85028	5,178	2,031	0.39	9
353	C83045	13,775	5,408	0.39	9
354	A83024	5,750	2,259	0.39	9
355	A83074	11,359	4,466	0.39	9
356	H82053	7,026	2,763	0.39	9
357	G85009	1,420	559	0.39	9
358	P88017	6,932	2,727	0.39	9
359	P86011	9,858	3,889	0.39	9
360	M88643	4,751	1,875	0.39	9
361	M88038	10,123	3,995	0.39	9
362	C83062	4,090	1,615	0.39	9
363	C83056	10,824	4,276	0.40	9
364	J83030	21,388	8,449	0.40	9
365	F81641	3,354	1,326	0.40	9
366	F81046	4,846	1,916	0.40	9
367	P89015	13,603	5,382	0.40	9
368	B87026	14,797	5,858	0.40	9
369	A83017	7,496	2,969	0.40	9
370	G82690	2,824	1,119	0.40	9
371	L81085	6,732	2,668	0.40	9
372	J83629	1,931	765	0.40	9
373	N83008	7,984	3,167	0.40	9
374	B83042	7,105	2,819	0.40	9
375	D83055	13,668	5,423	0.40	9
376	M87617	3,056	1,213	0.40	9
377	C83074	3,795	1,506	0.40	9
378	P83603	10,193	4,048	0.40	9
379	G82154	5,025	1,997	0.40	9
380	C83650	3,618	1,438	0.40	9
381	P92002	5,108	2,031	0.40	9
382	A84034	5,669	2,255	0.40	9
383	Y00683	1,584	631	0.40	9
384	L85065	2,281	909	0.40	9
385	P92016	16,682	6,657	0.40	9
386	M88008	5,370	2,144	0.40	9
387	B82017	14,273	5,700	0.40	9
388	M85694	2,418	966	0.40	9
389	J81074	5,678	2,269	0.40	9
390	C86037	10,539	4,212	0.40	9
391	J83609	2,602	1,040	0.40	9
392	N82067	3,393	1,356	0.40	9
393	F85010	4,528	1,811	0.40	9
394	H81642	3,023	1,211	0.40	9
395	A89623	2,500	1,002	0.40	9
396	B81025	12,266	4,916	0.40	9
397	Y03661	4,233	1,697	0.40	9
398	P86003	13,151	5,273	0.40	9
399	M88012	10,677	4,283	0.40	9
400	P89016	9,392	3,768	0.40	9
401	C83085	8,340	3,346	0.40	9
402	M84016	4,340	1,742	0.40	9
403	A84026	9,241	3,710	0.40	9
404	M83004	11,855	4,761	0.40	9
405	G82796	6,580	2,644	0.40	9
406	B82024	7,704	3,096	0.40	9
407	C84646	3,798	1,529	0.40	9
408	J81046	11,664	4,698	0.40	9
409	B85018	8,850	3,565	0.40	9
410	A83070	5,539	2,232	0.40	9
411	F81052	13,669	5,513	0.40	9
412	B86023	3,958	1,597	0.40	9
413	C83049	6,613	2,671	0.40	9
414	L85013	3,373	1,365	0.40	9
415	M83103	4,083	1,653	0.40	9
416	H81015	11,077	4,486	0.41	9
417	L85004	11,111	4,500	0.41	9
418	N81018	7,792	3,157	0.41	9
419	L85020	9,395	3,807	0.41	9
420	F81743	4,990	2,022	0.41	9
421	F86009	7,528	3,051	0.41	9
422	H85643	2,397	972	0.41	9
423	J84016	5,741	2,328	0.41	9
424	P92646	3,074	1,247	0.41	9
425	B87006	9,591	3,893	0.41	9
426	P86019	9,110	3,702	0.41	9
427	C83067	5,813	2,364	0.41	9
428	M91009	5,591	2,277	0.41	9
429	A85616	2,425	989	0.41	9
430	C88091	2,738	1,118	0.41	9
431	Y02633	2,716	1,109	0.41	9
432	G82764	1,416	578	0.41	9
433	B85015	8,985	3,674	0.41	9
434	M88616	7,462	3,052	0.41	9
435	M87030	3,885	1,590	0.41	9
436	M85048	11,459	4,690	0.41	9
437	P85005	9,651	3,952	0.41	9
438	M85158	2,302	943	0.41	9
439	L82620	5,318	2,181	0.41	9
440	E87718	90	37	0.41	9
441	B85640	3,523	1,446	0.41	9
442	D82053	5,474	2,248	0.41	9
443	E83613	2,747	1,128	0.41	9
444	C83065	6,818	2,806	0.41	9
445	M91616	7,593	3,126	0.41	9
446	P81136	2,493	1,028	0.41	9
447	N81006	5,019	2,072	0.41	9
448	E82082	13,988	5,776	0.41	9
449	P86606	4,454	1,841	0.41	9
450	L85032	6,013	2,485	0.41	9
451	J83059	7,011	2,902	0.41	9
452	H85054	4,944	2,047	0.41	9
453	C83058	6,886	2,855	0.41	9
454	B83604	2,776	1,153	0.42	9
455	P88023	6,506	2,707	0.42	9
456	N85001	4,090	1,703	0.42	9
457	D83001	10,649	4,437	0.42	9
458	B85620	2,008	837	0.42	9
459	F81212	7,886	3,288	0.42	9
460	B84002	6,257	2,609	0.42	9
461	H82021	6,375	2,663	0.42	9
462	P84047	4,715	1,970	0.42	9
463	C84125	3,402	1,422	0.42	9
464	J81028	10,215	4,274	0.42	9
465	M87015	8,810	3,686	0.42	9
466	P89017	5,758	2,410	0.42	9
467	G82666	9,073	3,798	0.42	9
468	C86012	5,316	2,229	0.42	9
469	B82045	4,263	1,788	0.42	9
470	E87045	2,959	1,242	0.42	9
471	B86101	3,329	1,397	0.42	9
472	A83073	5,049	2,120	0.42	9
473	C83033	3,338	1,402	0.42	9
474	C84717	2,971	1,249	0.42	9
475	L83647	2,166	911	0.42	9
476	N82036	3,908	1,643	0.42	9
477	G82652	2,584	1,088	0.42	9
478	G83066	8,537	3,595	0.42	9
479	A89617	2,224	937	0.42	9
480	C84074	3,759	1,583	0.42	9
481	P81764	2,311	974	0.42	9
482	E87756	3,353	1,413	0.42	9
483	G82063	8,180	3,450	0.42	9
484	C84101	2,738	1,155	0.42	9
485	D83045	5,142	2,169	0.42	9
486	C85013	9,905	4,180	0.42	9
487	P92623	1,923	812	0.42	9
488	B81097	1,518	641	0.42	9
489	A89013	5,277	2,228	0.42	9
490	L85037	3,892	1,645	0.42	9
491	H82004	4,570	1,933	0.42	9
492	B82068	3,185	1,350	0.42	9
493	D82005	9,389	3,983	0.42	9
494	N83612	3,009	1,276	0.42	9
495	C83029	5,922	2,512	0.42	9
496	A83617	2,699	1,145	0.42	9
497	G82187	4,715	2,004	0.43	9
498	P87014	4,126	1,758	0.43	9
499	M88612	4,305	1,837	0.43	9
500	H82013	11,760	5,019	0.43	9
501	B81063	6,707	2,863	0.43	9
502	P91029	5,354	2,286	0.43	9
503	C84052	6,021	2,573	0.43	9
504	M89001	3,159	1,351	0.43	9
505	L82056	7,859	3,361	0.43	9
506	E84056	1,737	743	0.43	9
507	F81104	4,351	1,863	0.43	9
508	M83682	5,902	2,527	0.43	9
509	L84006	6,596	2,824	0.43	9
510	P92643	1,528	654	0.43	9
511	F85036	3,624	1,553	0.43	9
512	L84068	3,911	1,678	0.43	9
513	A81001	4,234	1,817	0.43	9
514	B81120	1,682	722	0.43	9
515	P92031	3,361	1,445	0.43	9
516	C85019	4,905	2,111	0.43	9
517	L85036	5,493	2,368	0.43	9
518	B81642	2,864	1,235	0.43	9
519	C86005	5,870	2,533	0.43	9
520	J83011	14,100	6,084	0.43	9
521	B81069	9,922	4,289	0.43	9
522	C83015	8,310	3,596	0.43	9
523	E85045	3,706	1,604	0.43	9
524	P88012	8,517	3,686	0.43	9
525	C85614	2,811	1,218	0.43	9
526	A85014	7,307	3,167	0.43	9
527	C85617	3,354	1,454	0.43	9
528	C83083	7,869	3,414	0.43	9
529	M82025	7,531	3,268	0.43	9
530	A83057	6,279	2,726	0.43	9
531	L84031	5,460	2,372	0.43	9
532	P81687	1,642	714	0.43	9
533	G81662	3,749	1,631	0.44	9
534	G82218	5,652	2,459	0.44	9
535	C83025	9,080	3,955	0.44	9
536	P89029	5,846	2,546	0.44	9
537	M82026	12,748	5,570	0.44	9
538	E82626	2,589	1,132	0.44	9
539	L83626	2,134	934	0.44	9
540	G82114	14,245	6,248	0.44	9
541	B81099	5,582	2,449	0.44	9
542	A88608	5,721	2,510	0.44	9
543	B81053	11,959	5,248	0.44	9
544	B86658	4,351	1,912	0.44	9
545	L82035	7,421	3,268	0.44	9
546	M88630	5,586	2,461	0.44	9
547	M82038	9,734	4,291	0.44	9
548	M82046	3,822	1,685	0.44	9
549	M92001	3,400	1,500	0.44	9
550	M87007	9,407	4,156	0.44	9
551	G82160	10,962	4,843	0.44	9
552	P92021	6,238	2,758	0.44	9
553	C83022	19,807	8,757	0.44	9
554	G82013	13,124	5,806	0.44	9
555	N82034	6,064	2,684	0.44	9
556	M88016	1,915	848	0.44	9
557	G82679	6,782	3,003	0.44	9
558	M83072	11,678	5,176	0.44	9
559	H82007	11,892	5,273	0.44	9
560	A84032	2,452	1,089	0.44	9
561	P85010	9,593	4,279	0.45	9
562	G82219	4,558	2,037	0.45	9
563	E87005	7,166	3,204	0.45	9
564	M88003	11,388	5,095	0.45	9
565	P81774	2,111	945	0.45	9
566	A81025	4,162	1,863	0.45	9
567	B82619	1,097	491	0.45	9
568	G82791	4,253	1,905	0.45	9
569	N83026	2,618	1,174	0.45	9
570	G82112	7,275	3,264	0.45	9
571	C88084	3,386	1,519	0.45	9
572	D83014	9,543	4,291	0.45	9
573	G82733	3,222	1,449	0.45	9
574	P92020	4,218	1,899	0.45	9
575	F81209	2,281	1,027	0.45	9
576	C83011	9,142	4,119	0.45	9
577	P92042	4,471	2,016	0.45	9
578	C82013	9,767	4,406	0.45	9
579	B84017	2,201	994	0.45	9
580	G81039	14,892	6,724	0.45	9
581	G82184	1,637	740	0.45	9
582	M88024	3,045	1,378	0.45	9
583	C83032	7,572	3,429	0.45	9
584	P91008	6,175	2,797	0.45	9
585	P82014	7,463	3,381	0.45	9
586	J81048	6,175	2,803	0.45	9
587	G82762	1,582	718	0.45	9
588	L83021	9,891	4,492	0.45	9
589	A89604	3,938	1,790	0.45	9
590	B81039	17,420	7,943	0.46	9
591	J81066	9,962	4,546	0.46	9
592	F81121	6,605	3,028	0.46	9
593	G82097	16,794	7,701	0.46	9
594	L84011	7,027	3,238	0.46	9
595	C85005	8,082	3,724	0.46	9
596	C84142	4,947	2,280	0.46	9
597	E82008	9,244	4,261	0.46	9
598	C87006	9,112	4,201	0.46	9
599	M82006	9,962	4,616	0.46	9
600	P85013	16,784	7,787	0.46	9
601	J81054	9,754	4,539	0.47	9
602	J82210	9,258	4,309	0.47	9
603	P85008	13,010	6,065	0.47	9
604	F81157	5,463	2,547	0.47	9
605	C82656	2,232	1,043	0.47	9
606	D83033	9,004	4,209	0.47	9
607	G81052	4,062	1,902	0.47	9
608	B86673	3,561	1,673	0.47	9
609	H82087	4,080	1,918	0.47	9
610	M87018	5,936	2,792	0.47	9
611	D82065	6,833	3,218	0.47	9
612	G82077	9,195	4,331	0.47	9
613	A89620	2,662	1,255	0.47	9
614	C83040	16,014	7,565	0.47	9
615	P85020	7,670	3,629	0.47	9
616	C83652	2,130	1,008	0.47	9
617	M83090	3,438	1,629	0.47	9
618	J83601	3,483	1,655	0.48	9
619	P91625	4,899	2,332	0.48	9
620	C83613	5,499	2,619	0.48	9
621	B81648	1,237	591	0.48	9
622	L83639	2,214	1,061	0.48	9
623	B85026	8,072	3,872	0.48	9
624	J82166	10,587	5,079	0.48	9
625	G82073	9,542	4,581	0.48	9
626	C88036	9,654	4,649	0.48	9
627	G81051	6,894	3,328	0.48	9
628	B87030	10,019	4,840	0.48	9
629	F81670	2,993	1,447	0.48	9
630	L83045	14,411	6,992	0.49	9
631	C87031	3,249	1,577	0.49	9
632	F81032	2,612	1,268	0.49	9
633	M82031	2,914	1,415	0.49	9
634	C85017	8,170	3,968	0.49	9
635	M83121	6,076	2,958	0.49	9
636	P88009	3,482	1,700	0.49	9
637	C83031	11,398	5,582	0.49	9
638	L83643	1,849	906	0.49	9
639	H82014	9,612	4,713	0.49	9
640	J81002	8,573	4,207	0.49	9
641	B86024	9,056	4,444	0.49	9
642	C85003	11,029	5,418	0.49	9
643	M88030	6,268	3,088	0.49	9
644	J82027	5,550	2,744	0.49	9
645	P88010	1,528	758	0.50	9
646	C86609	3,083	1,531	0.50	9
647	G82691	2,988	1,485	0.50	9
648	P82011	5,995	2,980	0.50	9
649	P85006	6,068	3,018	0.50	9
650	Y02753	3,497	1,740	0.50	9
651	Y02625	1,095	546	0.50	9
652	C87013	6,200	3,099	0.50	9
653	M88001	6,666	3,335	0.50	9
654	C83042	5,710	2,863	0.50	9
655	B86056	5,399	2,707	0.50	9
656	M83017	3,989	2,001	0.50	9
657	C83064	13,954	7,005	0.50	9
658	G82124	4,057	2,037	0.50	9
659	E83639	3,558	1,787	0.50	9
660	C86021	6,374	3,210	0.50	9
661	B86033	11,247	5,666	0.50	9
662	M88022	4,414	2,225	0.50	9
663	P81107	7,703	3,937	0.51	9
664	A82065	2,939	1,506	0.51	9
665	A89007	10,134	5,201	0.51	9
666	J82007	13,305	6,862	0.52	9
667	M92027	3,275	1,702	0.52	9
668	C83019	22,401	11,680	0.52	9
669	C84650	1,764	920	0.52	9
670	B85019	16,953	8,859	0.52	9
671	G82032	13,760	7,213	0.52	9
672	P81087	11,184	5,876	0.53	9
673	P92635	2,464	1,302	0.53	9
674	G82226	4,640	2,456	0.53	9
675	F84713	2,358	1,249	0.53	9
676	M88035	3,014	1,604	0.53	9
677	M88007	9,728	5,219	0.54	9
678	C85026	6,190	3,328	0.54	9
679	G82057	10,806	5,823	0.54	9
680	G84006	11,281	6,106	0.54	9
681	L85003	3,850	2,090	0.54	9
682	P86620	2,164	1,176	0.54	9
683	A83632	2,507	1,369	0.55	9
684	C83055	5,131	2,807	0.55	9
685	M88023	3,706	2,038	0.55	9
686	L83088	4,348	2,392	0.55	9
687	F81151	4,473	2,461	0.55	9
688	N82077	2,781	1,540	0.55	9
689	F81711	1,588	880	0.55	9
690	L85624	2,255	1,251	0.55	9
691	P84049	10,108	5,614	0.56	9
692	B87036	3,996	2,229	0.56	9
693	F81697	3,193	1,785	0.56	9
694	P82609	6,117	3,440	0.56	9
695	Y04887	1,802	1,015	0.56	9
696	B81635	3,147	1,779	0.57	9
697	G82667	1,598	905	0.57	9
698	A83644	3,432	1,950	0.57	9
699	J81640	1,891	1,075	0.57	9
700	P92006	6,227	3,558	0.57	9
701	J82021	8,400	4,799	0.57	9
702	C83018	9,790	5,663	0.58	9
703	A83032	2,412	1,402	0.58	9
704	G82147	5,714	3,325	0.58	9
705	F81661	756	441	0.58	9
706	C88626	1,599	934	0.58	9
707	G82007	7,226	4,226	0.58	9
708	P85019	4,074	2,393	0.59	9
709	P85605	4,564	2,684	0.59	9
710	G82700	3,733	2,213	0.59	9
711	G82665	4,267	2,530	0.59	9
712	L82052	9,669	5,745	0.59	9
713	P88618	1,734	1,031	0.59	9
714	C83036	6,449	3,843	0.60	9
715	P85015	7,828	4,666	0.60	9
716	F81198	2,615	1,565	0.60	9
717	M88640	3,210	1,924	0.60	9
718	C86007	10,398	6,248	0.60	9
719	B81683	1,727	1,044	0.60	9
720	A89029	1,854	1,121	0.60	9
721	G82722	3,511	2,124	0.61	9
722	L82038	4,732	2,880	0.61	9
723	B81087	7,137	4,345	0.61	9
724	P92008	8,881	5,416	0.61	9
725	B81109	12,300	7,542	0.61	9
726	P91017	7,365	4,517	0.61	9
727	P82013	4,180	2,564	0.61	9
728	B85005	6,413	3,977	0.62	9
729	G82793	2,720	1,698	0.62	9
730	G82055	5,506	3,479	0.63	9
731	B81043	14,722	9,360	0.64	9
732	M88627	2,006	1,278	0.64	9
733	B83620	2,917	1,863	0.64	9
734	P84062	2,639	1,708	0.65	9
735	M84006	10,467	6,822	0.65	9
736	G82018	5,009	3,268	0.65	9
737	G82053	3,656	2,413	0.66	9
738	G81057	5,818	3,877	0.67	9
739	L83117	668	448	0.67	9
740	E83650	1,670	1,122	0.67	9
741	B87041	2,712	1,825	0.67	9
742	G82170	2,424	1,642	0.68	9
743	F81648	871	591	0.68	9
744	G82227	4,744	3,257	0.69	9
745	B86010	9,198	6,332	0.69	9
746	C85623	3,579	2,516	0.70	9
747	G82719	2,785	1,966	0.71	9
748	E83008	7,415	5,275	0.71	9
749	C88622	1,629	1,176	0.72	9
750	L83664	1,791	1,353	0.76	9
751	G81651	3,972	3,015	0.76	9
752	C84692	2,712	2,169	0.80	9
753	M81009	5,877	4,809	0.82	9
754	Y02516	807	665	0.82	9
755	H82641	2,098	1,729	0.82	9
756	E87727	157	130	0.83	9
757	P82020	4,005	3,345	0.84	9
758	E87694	257	225	0.88	9
759	M84031	6,702	6,009	0.90	9
760	M88646	2,102	2,047	0.97	9
761	F81633	3,363	3,300	0.98	9
762	E87711	284	295	1.04	9
763	K84066	388	459	1.18	9
764	Y04786	316	620	1.96	9
765	G82651	1	2	2.06	9
766	H84608	2,026	4,669	2.30	9
767	E87723	1	996	995.74	9

	PRACTICE	TOTAL NUMBER OF PATIENTS	TOTAL COST OF STATIN PRESCRIPTIONS (£)	STATIN COST PER PATIENT (£)
0	H84623	9,003	1,101	0.12
1	H83002	2,790	341	0.12
2	H84039	10,497	1,284	0.12
3	Y00568	3,460	423	0.12
4	A86003	32,431	3,965	0.12
5	E82603	2,401	293	0.12
6	M86613	2,925	357	0.12
7	B86019	9,218	1,126	0.12
8	F83682	1,791	219	0.12
9	H81029	12,064	1,473	0.12
10	Y00902	3,880	474	0.12
11	H83020	10,633	1,297	0.12
12	M85684	1,882	229	0.12
13	E87772	529	64	0.12
14	F84670	4,195	511	0.12
15	C81019	8,099	987	0.12
16	H84057	2,608	318	0.12
17	E82663	4,122	502	0.12
18	F84673	2,030	247	0.12
19	F85008	14,828	1,806	0.12
20	G85096	5,103	621	0.12
21	P81213	16,855	2,052	0.12
22	E87722	4,653	567	0.12
23	L85023	13,998	1,704	0.12
24	D81003	9,852	1,198	0.12
25	K83616	5,425	660	0.12
26	H83001	12,222	1,486	0.12
27	H85092	5,751	699	0.12
28	K81638	2,187	266	0.12
29	F85028	8,243	1,001	0.12
30	K81633	8,221	998	0.12
31	E84021	11,783	1,429	0.12
32	K84009	15,338	1,859	0.12
33	F84060	4,901	594	0.12
34	J82023	13,176	1,594	0.12
35	H82064	9,898	1,197	0.12
36	E84080	8,215	993	0.12
37	H84015	7,294	882	0.12
38	D81020	12,626	1,526	0.12
39	F82676	7,578	916	0.12
40	E85605	7,199	870	0.12
41	J82054	8,284	1,001	0.12
42	E84028	4,664	563	0.12
43	F84660	2,095	253	0.12
44	D81012	11,281	1,359	0.12
45	Y02638	1,775	214	0.12
46	E87069	2,837	342	0.12
47	E86042	6,545	788	0.12
48	G82647	5,906	711	0.12
49	H84055	11,117	1,337	0.12
50	H84023	9,045	1,086	0.12
51	Y00026	2,025	243	0.12
52	E84012	7,060	847	0.12
53	K84618	6,909	829	0.12
54	H84002	12,920	1,549	0.12
55	E85014	9,313	1,117	0.12
56	Y00228	2,226	267	0.12
57	Y00454	2,206	264	0.12
58	F81031	7,353	880	0.12
59	D81084	7,811	935	0.12
60	H81048	9,656	1,156	0.12
61	E81075	8,234	985	0.12
62	K81065	5,625	673	0.12
63	F86034	5,519	660	0.12
64	C81104	1,794	215	0.12
65	D81060	6,592	788	0.12
66	L81071	12,245	1,463	0.12
67	G85013	11,230	1,342	0.12
68	E84067	8,197	978	0.12
69	K81047	26,790	3,196	0.12
70	L85612	4,635	552	0.12
71	B86069	12,955	1,543	0.12
72	C82623	3,365	400	0.12
73	C81648	4,159	495	0.12
74	E84701	4,063	483	0.12
75	E84042	7,879	937	0.12
76	G85133	7,406	880	0.12
77	Y03506	5,234	622	0.12
78	F83060	8,274	983	0.12
79	J81072	12,746	1,514	0.12
80	H83015	13,232	1,572	0.12
81	G83068	1,041	124	0.12
82	C82094	6,618	785	0.12
83	D81082	7,189	853	0.12
84	E85098	3,556	421	0.12
85	L81622	4,037	478	0.12
86	G85632	6,817	807	0.12
87	E84680	4,025	476	0.12
88	E84635	4,681	554	0.12
89	F86705	2,816	333	0.12
90	F86657	7,589	898	0.12
91	D81017	5,663	670	0.12
92	Y02811	9,070	1,073	0.12
93	P84676	4,338	513	0.12
94	K82010	9,689	1,146	0.12
95	G85006	10,557	1,248	0.12
96	H84012	11,391	1,346	0.12
97	M81617	4,483	530	0.12
98	K84060	4,530	535	0.12
99	F82045	3,008	355	0.12
100	C83075	7,071	834	0.12
101	G85030	5,520	651	0.12
102	Y02423	36,448	4,296	0.12
103	G85055	12,398	1,461	0.12
104	K81060	10,359	1,220	0.12
105	E82054	11,349	1,336	0.12
106	F84035	3,677	433	0.12
107	G82011	6,042	711	0.12
108	B84613	8,315	978	0.12
109	F85019	12,513	1,470	0.12
110	E87043	6,295	739	0.12
111	C82610	5,219	613	0.12
112	J82050	16,335	1,917	0.12
113	K81082	13,708	1,608	0.12
114	F86607	14,401	1,689	0.12
115	E81611	3,220	377	0.12
116	B86108	4,894	574	0.12
117	H84639	4,323	507	0.12
118	C82093	11,170	1,309	0.12
119	Y02816	3,119	365	0.12
120	Y02826	10,277	1,201	0.12
121	P81735	2,852	333	0.12
122	Y03363	2,008	234	0.12
123	J82144	24,579	2,867	0.12
124	E87007	7,971	930	0.12
125	B84005	9,934	1,158	0.12
126	F84079	9,575	1,116	0.12
127	E82094	9,870	1,150	0.12
128	D81081	3,091	360	0.12
129	F85063	12,624	1,470	0.12
130	F81665	4,255	496	0.12
131	H83007	10,494	1,222	0.12
132	A86007	3,101	361	0.12
133	N82079	6,483	754	0.12
134	K82076	13,077	1,522	0.12
135	G81045	10,047	1,169	0.12
136	Y02747	11,415	1,328	0.12
137	Y00078	3,885	451	0.12
138	F84105	10,782	1,252	0.12
139	E84674	6,391	740	0.12
140	K82610	6,819	789	0.12
141	H84005	11,111	1,285	0.12
142	G81018	11,234	1,299	0.12
143	H85634	7,528	870	0.12
144	E82053	6,786	784	0.12
145	Y02833	3,889	449	0.12
146	E84013	8,023	926	0.12
147	C88062	4,535	524	0.12
148	E86620	4,534	523	0.12
149	H82088	7,603	877	0.12
150	E82655	3,195	368	0.12
151	Y02896	1,511	174	0.12
152	G85020	4,714	543	0.12
153	E86003	4,036	465	0.12
154	D82632	3,053	351	0.11
155	K84062	6,911	793	0.11
156	Y02710	3,649	419	0.11
157	Y00028	7,175	823	0.11
158	H83006	5,883	675	0.11
159	F86686	2,371	272	0.11
160	F84749	11,883	1,361	0.11
161	G85623	9,061	1,038	0.11
162	F83660	8,806	1,008	0.11
163	C84602	2,252	258	0.11
164	D81031	15,281	1,748	0.11
165	F84050	9,538	1,090	0.11
166	G85642	6,594	753	0.11
167	H84048	7,287	832	0.11
168	C81033	15,788	1,802	0.11
169	Y02519	5,378	613	0.11
170	F82034	4,834	551	0.11
171	E87010	6,762	770	0.11
172	K83009	11,390	1,296	0.11
173	P84689	4,845	551	0.11
174	F84055	3,654	415	0.11
175	E86625	3,313	377	0.11
176	K84011	16,003	1,818	0.11
177	G85073	7,618	864	0.11
178	K84080	4,901	556	0.11
179	E87753	7,265	824	0.11
180	M85145	5,902	669	0.11
181	G85052	7,012	795	0.11
182	H84006	9,658	1,094	0.11
183	E87720	7,318	829	0.11
184	G85130	8,737	989	0.11
185	H83039	7,033	795	0.11
186	B83617	1,994	225	0.11
187	L81108	1,709	193	0.11
188	E84601	7,221	815	0.11
189	K84073	7,599	858	0.11
190	F83674	3,018	340	0.11
191	F83023	20,284	2,287	0.11
192	F84601	5,720	645	0.11
193	A83011	28,350	3,191	0.11
194	N82076	6,547	736	0.11
195	G81103	4,100	461	0.11
196	F86692	6,440	724	0.11
197	F81073	3,722	418	0.11
198	Y02838	4,617	518	0.11
199	N82039	3,027	339	0.11
200	Y02586	3,018	338	0.11
201	E84693	7,959	891	0.11
202	B85044	2,425	272	0.11
203	B86103	2,283	256	0.11
204	D81062	7,217	808	0.11
205	C81653	4,146	464	0.11
206	E85116	5,056	566	0.11
207	G85038	13,435	1,502	0.11
208	M85064	4,271	477	0.11
209	Y02520	5,564	621	0.11
210	E85006	4,502	502	0.11
211	G81047	7,840	875	0.11
212	F82639	2,548	284	0.11
213	K84621	4,253	474	0.11
214	J81024	10,271	1,144	0.11
215	G81090	6,073	676	0.11
216	C82119	2,457	274	0.11
217	K84075	9,901	1,102	0.11
218	F83021	13,305	1,480	0.11
219	H81647	2,341	260	0.11
220	H83050	5,980	664	0.11
221	B83614	5,452	605	0.11
222	J82629	2,784	309	0.11
223	P81756	3,395	376	0.11
224	F83635	3,514	389	0.11
225	H84044	8,821	976	0.11
226	N81628	8,665	958	0.11
227	C81633	1,758	194	0.11
228	F86086	5,673	626	0.11
229	A82623	5,271	580	0.11
230	B84016	8,089	890	0.11
231	H81010	23,717	2,608	0.11
232	G85692	10,724	1,179	0.11
233	F81014	11,444	1,255	0.11
234	J82079	6,451	707	0.11
235	F85013	10,214	1,119	0.11
236	F82653	2,578	282	0.11
237	G85050	3,829	418	0.11
238	L81078	13,700	1,497	0.11
239	G85084	6,538	713	0.11
240	D81616	4,402	480	0.11
241	E84022	6,904	752	0.11
242	K82631	14,141	1,539	0.11
243	E86010	11,337	1,233	0.11
244	F84119	10,991	1,195	0.11
245	H85111	10,014	1,088	0.11
246	H84637	2,331	253	0.11
247	L83016	35,703	3,875	0.11
248	J81033	18,362	1,992	0.11
249	K84065	2,678	289	0.11
250	F84022	6,404	692	0.11
251	K81074	15,159	1,636	0.11
252	H85051	16,434	1,774	0.11
253	G85088	6,184	666	0.11
254	Y00507	2,275	245	0.11
255	F84033	12,276	1,322	0.11
256	Y00561	2,350	253	0.11
257	C82667	6,937	747	0.11
258	Y00079	1,996	215	0.11
259	F84740	4,713	506	0.11
260	F83044	4,294	461	0.11
261	P92620	3,988	428	0.11
262	F81669	4,927	529	0.11
263	G85134	5,578	599	0.11
264	K82621	5,566	597	0.11
265	F83671	2,741	294	0.11
266	G85123	7,143	765	0.11
267	Y00020	5,963	639	0.11
268	E84002	6,626	709	0.11
269	C82662	3,476	372	0.11
270	E82658	4,221	451	0.11
271	G85015	10,617	1,134	0.11
272	M83726	3,957	423	0.11
273	E82638	6,747	720	0.11
274	Y02464	2,675	285	0.11
275	E83637	6,836	727	0.11
276	G85109	15,450	1,642	0.11
277	M85735	5,020	534	0.11
278	C81652	4,012	426	0.11
279	E82656	2,355	250	0.11
280	J82081	23,546	2,498	0.11
281	Y01050	6,660	706	0.11
282	G85651	9,942	1,054	0.11
283	Y04809	2,259	239	0.11
284	G83067	26,312	2,786	0.11
285	E81633	2,907	308	0.11
286	E85677	4,505	477	0.11
287	G85712	3,145	333	0.11
288	C81110	10,411	1,101	0.11
289	H83044	13,789	1,458	0.11
290	F86045	1,914	202	0.11
291	H85012	25,397	2,672	0.11
292	H83011	4,290	451	0.11
293	A82045	8,352	878	0.11
294	F85615	4,903	515	0.11
295	F83010	16,903	1,772	0.10
296	E82057	5,344	560	0.10
297	G85091	4,466	467	0.10
298	H83017	16,985	1,777	0.10
299	F84117	9,314	974	0.10
300	G85135	12,101	1,263	0.10
301	D81619	4,822	502	0.10
302	M83127	4,665	486	0.10
303	Y00212	4,881	508	0.10
304	G85673	6,643	692	0.10
305	F84114	7,575	789	0.10
306	Y02974	6,820	710	0.10
307	E84663	6,315	657	0.10
308	K83619	4,113	428	0.10
309	H85009	11,953	1,242	0.10
310	Y03035	19,499	2,025	0.10
311	J82071	9,150	950	0.10
312	Y02512	2,318	241	0.10
313	G81028	13,717	1,423	0.10
314	E82060	19,971	2,070	0.10
315	M82050	8,124	841	0.10
316	E85746	3,581	370	0.10
317	C81665	10,570	1,092	0.10
318	E84708	2,248	232	0.10
319	Y00033	5,645	582	0.10
320	E87742	2,530	261	0.10
321	E85628	8,641	889	0.10
322	F84025	4,689	482	0.10
323	C84664	3,725	383	0.10
324	K81647	2,557	263	0.10
325	H83030	3,188	327	0.10
326	G81694	2,210	227	0.10
327	Y00185	9,723	997	0.10
328	F85071	7,865	805	0.10
329	P87651	4,885	499	0.10
330	F83032	10,572	1,078	0.10
331	Y01165	3,711	378	0.10
332	H85100	7,236	737	0.10
333	Y02757	5,942	603	0.10
334	G85047	5,440	551	0.10
335	F85023	5,735	581	0.10
336	K84032	9,120	922	0.10
337	K81092	12,022	1,216	0.10
338	E82014	12,680	1,282	0.10
339	F83039	5,350	540	0.10
340	D81063	11,724	1,183	0.10
341	E87762	3,428	346	0.10
342	G81034	25,138	2,527	0.10
343	G85698	8,536	857	0.10
344	F85067	4,748	477	0.10
345	F83015	12,502	1,254	0.10
346	M92028	9,729	975	0.10
347	B81675	7,865	787	0.10
348	G84041	2,994	298	0.10
349	C81627	4,051	403	0.10
350	H85683	6,108	608	0.10
351	F83061	4,029	401	0.10
352	Y00186	1,436	143	0.10
353	H84027	18,068	1,796	0.10
354	H85680	10,615	1,055	0.10
355	J81625	7,847	779	0.10
356	M82045	4,681	464	0.10
357	F83059	17,486	1,732	0.10
358	K81640	2,398	237	0.10
359	K84079	9,069	896	0.10
360	Y02471	7,796	770	0.10
361	G85100	6,305	622	0.10
362	Y02164	6,692	660	0.10
363	Y02581	3,318	326	0.10
364	F85700	3,693	363	0.10
365	M86029	17,247	1,693	0.10
366	F84092	6,247	613	0.10
367	E86609	3,268	320	0.10
368	F83063	11,427	1,117	0.10
369	J82220	6,562	641	0.10
370	E81615	3,335	326	0.10
371	C81072	22,523	2,196	0.10
372	M86004	18,081	1,761	0.10
373	Y02483	4,947	481	0.10
374	Y00527	1,942	189	0.10
375	G81044	6,540	635	0.10
376	G85025	10,728	1,041	0.10
377	F86708	2,929	284	0.10
378	E85122	15,928	1,541	0.10
379	F85031	9,660	934	0.10
380	M85730	3,132	303	0.10
381	Y00352	5,138	497	0.10
382	Y03049	7,298	705	0.10
383	G85094	16,477	1,591	0.10
384	L81061	8,314	802	0.10
385	Y00092	2,449	236	0.10
386	N81651	3,102	298	0.10
387	Y01068	4,728	455	0.10
388	E83658	2,436	234	0.10
389	E82644	2,084	200	0.10
390	F84021	9,857	943	0.10
391	Y02515	25	2	0.10
392	E85091	1,700	162	0.10
393	F84716	4,702	449	0.10
394	D81625	18,725	1,782	0.10
395	F83664	6,980	664	0.10
396	E84025	14,210	1,348	0.09
397	E84033	4,886	463	0.09
398	G85132	7,092	672	0.09
399	F83665	12,011	1,139	0.09
400	F82661	3,508	332	0.09
401	G85674	5,358	507	0.09
402	F85069	7,108	670	0.09
403	D81002	15,717	1,481	0.09
404	H83627	6,179	582	0.09
405	Y02961	5,357	504	0.09
406	K82035	3,320	312	0.09
407	G85044	5,632	528	0.09
408	E87714	4,087	383	0.09
409	M86622	2,784	261	0.09
410	G85138	13,340	1,247	0.09
411	G85087	6,666	623	0.09
412	H85069	12,810	1,197	0.09
413	E85692	7,359	685	0.09
414	F83022	14,135	1,311	0.09
415	N82617	5,279	489	0.09
416	M85002	17,814	1,649	0.09
417	H83012	9,666	894	0.09
418	E84696	1,681	155	0.09
419	M91029	3,734	345	0.09
420	M92041	4,599	424	0.09
421	Y02002	13,052	1,203	0.09
422	Y03135	9,722	896	0.09
423	E85117	3,370	310	0.09
424	F85705	11,665	1,073	0.09
425	E85718	4,647	427	0.09
426	D81620	1,861	171	0.09
427	B86086	7,070	649	0.09
428	E84637	2,958	272	0.09
429	F83624	10,592	969	0.09
430	P84683	6,933	634	0.09
431	K84021	8,408	769	0.09
432	F84093	16,459	1,505	0.09
433	E82031	18,353	1,677	0.09
434	G81094	4,552	416	0.09
435	G85708	9,603	877	0.09
436	Y02960	5,285	482	0.09
437	F84632	939	86	0.09
438	F83045	10,835	988	0.09
439	G85016	10,098	921	0.09
440	M85041	9,198	838	0.09
441	F84717	9,145	833	0.09
442	E84032	8,850	805	0.09
443	E85687	1,138	104	0.09
444	B83602	6,710	610	0.09
445	L81632	10,761	976	0.09
446	J82082	10,884	987	0.09
447	Y02686	3,444	312	0.09
448	F84730	5,804	525	0.09
449	L81112	11,142	1,006	0.09
450	N82059	4,668	421	0.09
451	D81633	9,630	869	0.09
452	K81657	4,264	384	0.09
453	D81606	4,204	378	0.09
454	G85690	6,099	549	0.09
455	E84063	5,889	528	0.09
456	Y02646	6,069	544	0.09
457	H84041	3,925	349	0.09
458	F84636	3,368	299	0.09
459	E84705	2,493	221	0.09
460	H84061	9,011	798	0.09
461	H83623	2,554	226	0.09
462	Y02787	5,384	476	0.09
463	C82060	3,832	339	0.09
464	F86626	5,199	458	0.09
465	K81656	4,943	436	0.09
466	G85019	3,813	336	0.09
467	J81645	2,942	259	0.09
468	K82615	10,538	927	0.09
469	F84668	6,898	606	0.09
470	Y00486	7,100	623	0.09
471	B85634	3,268	286	0.09
472	K84049	6,615	578	0.09
473	Y02466	8,787	767	0.09
474	F84747	18,992	1,657	0.09
475	F84700	907	79	0.09
476	E85715	4,232	369	0.09
477	H83631	9,260	804	0.09
478	F83652	9,940	861	0.09
479	B86682	5,049	437	0.09
480	Y04969	4,259	369	0.09
481	G85711	3,846	333	0.09
482	F81141	13,879	1,201	0.09
483	F81632	3,738	323	0.09
484	Y01066	5,886	508	0.09
485	G81038	17,576	1,516	0.09
486	G85085	8,745	754	0.09
487	B86623	3,495	300	0.09
488	Y02973	5,793	496	0.09
489	K84076	4,897	418	0.09
490	M85085	2,050	174	0.09
491	F81192	9,570	814	0.09
492	F83666	5,575	474	0.08
493	E87063	10,760	914	0.08
494	M85021	2,733	232	0.08
495	C84621	4,279	362	0.08
496	G85105	5,271	446	0.08
497	Y02671	14,131	1,193	0.08
498	M86009	20,914	1,762	0.08
499	Y02589	7,435	626	0.08
500	M85722	5,249	442	0.08
501	C84691	10,046	841	0.08
502	Y02794	6,869	574	0.08
503	G85011	16,408	1,368	0.08
504	F86704	1,762	147	0.08
505	H83051	7,391	614	0.08
506	Y04965	2,859	238	0.08
507	G81638	9,998	828	0.08
508	Y02596	3,128	258	0.08
509	F84686	6,917	570	0.08
510	F84123	12,049	993	0.08
511	B82104	7,235	594	0.08
512	H84020	22,008	1,800	0.08
513	Y00206	5,048	412	0.08
514	H83037	7,027	573	0.08
515	Y02674	658	54	0.08
516	Y03063	7,125	579	0.08
517	G81703	16	1	0.08
518	E85617	3,235	262	0.08
519	K84617	3,880	313	0.08
520	D81001	10,932	877	0.08
521	H83624	7,106	568	0.08
522	C84628	5,392	430	0.08
523	C81091	9,176	732	0.08
524	Y02672	8,172	651	0.08
525	F85707	3,999	318	0.08
526	Y02469	4,900	389	0.08
527	G85137	7,676	609	0.08
528	A81621	2,245	178	0.08
529	Y02684	2,709	214	0.08
530	H85659	4,279	338	0.08
531	Y02755	4,384	346	0.08
532	K84026	11,144	879	0.08
533	Y01177	7,972	627	0.08
534	Y02597	2,743	216	0.08
535	Y02736	4,993	392	0.08
536	P84669	10,893	851	0.08
537	Y00243	573	45	0.08
538	C82105	2,703	211	0.08
539	G85112	4,554	355	0.08
540	D81056	7,146	557	0.08
541	Y02968	5,850	453	0.08
542	F85682	4,701	364	0.08
543	L81132	4,457	345	0.08
544	B83628	4,435	342	0.08
545	H83005	11,739	905	0.08
546	J83063	12,503	963	0.08
547	C87622	2,068	159	0.08
548	Y05212	1,808	138	0.08
549	F84635	7,741	591	0.08
550	Y00054	77	6	0.08
551	G85045	8,574	653	0.08
552	H83608	4,078	310	0.08
553	M85784	1,118	85	0.08
554	Y02906	4,104	310	0.08
555	Y00060	521	39	0.08
556	Y00184	8,303	625	0.08
557	D81631	11,824	886	0.07
558	G81020	2,107	158	0.07
559	B81692	2,920	218	0.07
560	K84624	9,555	710	0.07
561	H83034	7,844	580	0.07
562	P81146	5,134	377	0.07
563	D81630	8,834	647	0.07
564	E84667	2,597	190	0.07
565	Y02842	5,054	369	0.07
566	M81087	3,881	283	0.07
567	P84072	20,530	1,495	0.07
568	F84013	13,980	1,018	0.07
569	G85136	10,388	754	0.07
570	Y01962	9,797	711	0.07
571	B81666	2,209	160	0.07
572	Y02688	5,703	412	0.07
573	K84016	14,492	1,046	0.07
574	Y01924	112	8	0.07
575	E84646	6,606	475	0.07
576	J82024	4,970	357	0.07
577	Y02890	4,880	350	0.07
578	K81100	5,190	372	0.07
579	Y02807	7,715	552	0.07
580	L81012	18,263	1,303	0.07
581	K84063	7,975	568	0.07
582	F84685	3,079	219	0.07
583	D81070	9,071	641	0.07
584	Y02319	4,608	324	0.07
585	Y02342	8,287	582	0.07
586	H85049	7,386	517	0.07
587	E85013	3,504	245	0.07
588	E85034	6,160	428	0.07
589	D81013	14,901	1,035	0.07
590	L85616	4,418	307	0.07
591	H85075	3,999	277	0.07
592	C82670	1,089	75	0.07
593	Y02769	5,100	350	0.07
594	G85081	4,032	274	0.07
595	Y01280	2,366	160	0.07
596	E87741	3,879	261	0.07
597	Y00403	5,468	368	0.07
598	B81032	3,134	210	0.07
599	B83657	3,595	241	0.07
600	Y02928	7,943	532	0.07
601	P84018	6,473	426	0.07
602	F83686	6,051	397	0.07
603	G83065	15,793	1,036	0.07
604	D81054	19,392	1,267	0.07
605	G85695	5,013	325	0.06
606	G85076	5,566	361	0.06
607	Y02778	5,460	354	0.06
608	P81064	18,971	1,217	0.06
609	Y02986	2,404	153	0.06
610	P84630	14,619	925	0.06
611	M85680	2,844	179	0.06
612	Y02812	8,504	536	0.06
613	F84692	6,029	379	0.06
614	Y00265	6,702	421	0.06
615	E86004	9,623	602	0.06
616	Y02636	3,668	229	0.06
617	G85647	6,478	400	0.06
618	N81655	347	21	0.06
619	B85659	1,700	105	0.06
620	B86669	1,315	81	0.06
621	M85721	6,614	406	0.06
622	P81152	5,869	359	0.06
623	D81005	13,671	828	0.06
624	K84078	7,165	433	0.06
625	A82613	967	58	0.06
626	H85066	18,937	1,129	0.06
627	F83005	8,503	504	0.06
628	L83673	579	34	0.06
629	K83621	8,428	496	0.06
630	J82663	4,581	268	0.06
631	F85708	858	50	0.06
632	F85669	11,701	682	0.06
633	G85002	14,468	842	0.06
634	F81742	3,123	181	0.06
635	H85001	16,292	913	0.06
636	Y02639	7,973	434	0.05
637	E87740	1,763	94	0.05
638	Y02754	5,482	290	0.05
639	M91016	6,200	328	0.05
640	H84619	10,808	568	0.05
641	G85127	5,756	301	0.05
642	Y02566	4,933	257	0.05
643	Y00492	6,463	335	0.05
644	Y02476	8,683	447	0.05
645	E85686	3,889	199	0.05
646	F82670	2,325	119	0.05
647	Y02692	9,467	480	0.05
648	E87637	10,083	511	0.05
649	Y02751	9,525	482	0.05
650	Y02900	9,272	465	0.05
651	Y02494	4,332	217	0.05
652	Y02421	3,039	152	0.05
653	G81689	1,381	69	0.05
654	N82117	35,236	1,742	0.05
655	M87634	1,294	63	0.05
656	Y00049	172	8	0.05
657	D81637	10,486	506	0.05
658	Y02962	5,711	272	0.05
659	Y02526	6,271	299	0.05
660	Y03296	5,401	257	0.05
661	A81632	664	32	0.05
662	F85628	3,596	171	0.05
663	Y01964	6,063	284	0.05
664	F83058	11,446	537	0.05
665	Y02404	1,398	65	0.05
666	Y02442	4,653	217	0.05
667	Y03441	4,524	210	0.05
668	E84029	2,587	118	0.05
669	F84072	9,097	415	0.05
670	Y02957	5,417	246	0.05
671	G85724	13,098	595	0.05
672	Y04968	4,408	198	0.04
673	D81037	9,885	429	0.04
674	K84605	4,741	205	0.04
675	B82047	21,787	929	0.04
676	Y02823	3,933	167	0.04
677	G81646	3,158	133	0.04
678	B86651	1,875	78	0.04
679	K81605	19,765	801	0.04
680	P81762	3,357	135	0.04
681	Y00400	76	3	0.04
682	M85782	2,609	102	0.04
683	Y04273	8,456	329	0.04
684	C88007	28,980	1,110	0.04
685	B86102	311	12	0.04
686	K84013	20,611	775	0.04
687	H83634	1,659	62	0.04
688	G85715	6,422	233	0.04
689	Y02379	1,320	47	0.04
690	A89621	2,754	95	0.03
691	E82620	813	27	0.03
692	Y05167	10,249	336	0.03
693	M85778	11,529	375	0.03
694	M85167	20,180	656	0.03
695	Y02585	8,110	262	0.03
696	F83043	16,222	522	0.03
697	P91605	65	2	0.03
698	Y02463	7,676	236	0.03
699	Y04662	892	27	0.03
700	B81104	6,192	182	0.03
701	Y01655	9,104	267	0.03
702	H85664	2,492	72	0.03
703	Y04951	5,135	147	0.03
704	C84117	17,122	470	0.03
705	F81756	2,351	62	0.03
706	Y02847	9,107	232	0.03
707	J82217	456	11	0.02
708	E87677	15,696	374	0.02
709	E86030	11,860	281	0.02
710	Y02622	6,960	160	0.02
711	Y00996	2,195	49	0.02
712	E85724	920	20	0.02
713	A81630	712	15	0.02
714	F83672	11,026	234	0.02
715	Y01057	912	19	0.02
716	C84720	42	1	0.02
717	A81634	1,533	32	0.02
718	Y02578	7,808	164	0.02
719	C82020	18,445	376	0.02
720	Y00334	5,020	101	0.02
721	Y04225	849	17	0.02
722	Y00344	1,477	29	0.02
723	Y02676	6,348	123	0.02
724	Y02849	8,203	157	0.02
725	M83670	7,199	132	0.02
726	F84727	1,847	33	0.02
727	E85657	9,788	172	0.02
728	J82080	16,965	297	0.02
729	Y02946	7,451	126	0.02
730	G82140	16,551	250	0.02
731	A86027	14,111	201	0.01
732	C82124	19,940	260	0.01
733	A81633	917	12	0.01
734	C84714	6,896	79	0.01
735	J82199	17,445	198	0.01
736	Y02611	2,393	24	0.01
737	B85062	12,627	114	0.01
738	A84025	5,296	48	0.01
739	L81617	12,215	110	0.01
740	D83030	4,564	35	0.01
741	C84023	38,866	299	0.01
742	D82088	19,307	137	0.01
743	Y02999	837	5	0.01
744	B81057	2,573	15	0.01
745	E81626	5,582	33	0.01
746	Y02620	5,133	28	0.01
747	A83068	164	1	0.01
748	F84642	4,955	26	0.01
749	G81071	18,083	90	0.00
750	C82111	16,316	71	0.00
751	N82647	1,883	8	0.00
752	F83056	4,589	18	0.00
753	L85602	1,499	5	0.00
754	Y02988	325	1	0.00
755	J82096	6,276	19	0.00
756	B83051	9,577	25	0.00
757	Y02575	7,737	19	0.00
758	Y02795	2,924	4	0.00
759	N85643	1,779	3	0.00
760	F81720	3,272	5	0.00
761	N81607	2,676	3	0.00
762	M82612	2,308	2	0.00
763	P88633	3,239	3	0.00
764	E87768	10,971	8	0.00
765	Y03671	3,543	3	0.00
766	B86110	39,138	20	0.00
767	C88627	29,880	13	0.00

	count	mean	std	min	25%	50%	75%	max
First Decile	768	13,806	9,185	43	5,458	12,928	21,522	32,639
Last Decile	768	12,266	8,901	27	5,026	9,925	18,880	32,837

	REGION	MORTALITY RATES
0	East Midlands	0.0739%
1	East of England	0.0632%
2	London	0.0492%
3	North East	0.0834%
4	North West	0.0821%
5	South East	0.0571%
6	South West	0.0672%
7	West Midlands	0.0717%
8	Yorkshire and The Humber	0.0762%

	PRACTICE	MONTHLY PRACTICE SPEND (£)
0	A81001	83,221
1	A81002	333,515
2	A81003	47,977
3	A81004	147,495
4	A81005	107,670
5	A81006	197,437
6	A81007	141,781
7	A81008	62,464
8	A81009	139,613
9	A81011	188,196
10	A81012	67,075
11	A81013	93,712
12	A81014	72,566
13	A81015	51,006
14	A81016	135,797
15	A81017	275,156
16	A81018	84,741
17	A81019	169,984
18	A81020	132,334
19	A81021	245,132
20	A81022	168,050
21	A81023	106,771
22	A81025	73,392
23	A81026	236,931
24	A81027	193,919
25	A81029	76,445
26	A81030	212,856
27	A81031	209,266
28	A81032	171,345
29	A81033	60,723
30	A81034	321,551
31	A81035	182,831
32	A81036	267,374
33	A81037	118,084
34	A81038	55,752
35	A81039	124,855
36	A81040	124,912
37	A81041	129,426
38	A81042	103,079
39	A81043	134,543

	PRACTICE	MONTHLY PRACTICE SPEND (£)	TOTAL NUMBER OF PATIENTS	COST PER PATIENT (£)
0	A81001	83,221	4,234	19.66
1	A81002	333,515	19,691	16.94
2	A81003	47,977	3,357	14.29
3	A81004	147,495	9,151	16.12
4	A81005	107,670	7,883	13.66
5	A81006	197,437	12,345	15.99
6	A81007	141,781	9,744	14.55
7	A81008	62,464	4,027	15.51
8	A81009	139,613	9,175	15.22
9	A81011	188,196	11,435	16.46
10	A81012	67,075	4,749	14.12
11	A81013	93,712	6,182	15.16
12	A81014	72,566	3,821	18.99
13	A81015	51,006	3,060	16.67
14	A81016	135,797	9,630	14.10
15	A81017	275,156	20,746	13.26
16	A81018	84,741	5,473	15.48
17	A81019	169,984	10,381	16.37
18	A81020	132,334	7,907	16.74
19	A81021	245,132	13,189	18.59
20	A81022	168,050	10,743	15.64
21	A81023	106,771	7,674	13.91
22	A81025	73,392	4,162	17.63
23	A81026	236,931	13,758	17.22
24	A81027	193,919	13,255	14.63
25	A81029	76,445	5,799	13.18
26	A81030	212,856	12,119	17.56
27	A81031	209,266	12,729	16.44
28	A81032	171,345	10,505	16.31
29	A81033	60,723	3,371	18.01
30	A81034	321,551	21,774	14.77
31	A81035	182,831	10,751	17.01
32	A81036	267,374	17,362	15.40
33	A81037	118,084	7,045	16.76
34	A81038	55,752	3,311	16.84
35	A81039	124,855	9,359	13.34
36	A81040	124,912	8,552	14.61
37	A81041	129,426	9,088	14.24
38	A81042	103,079	6,379	16.16
39	A81043	134,543	8,002	16.81

	PRACTICE	TOTAL NUMBER OF PATIENTS	TOTAL COST OF STATIN PRESCRIPTIONS (£)	STATIN COST PER PATIENT (£)
0	A81001	4,234	1,817	0.43
1	A81002	19,691	6,514	0.33
2	A81003	3,357	868	0.26
3	A81004	9,151	2,135	0.23
4	A81005	7,883	1,887	0.24
5	A81006	12,345	4,075	0.33
6	A81007	9,744	2,246	0.23
7	A81008	4,027	1,398	0.35
8	A81009	9,175	1,588	0.17
9	A81011	11,435	2,946	0.26
10	A81012	4,749	950	0.20
11	A81013	6,182	1,114	0.18
12	A81014	3,821	1,345	0.35
13	A81015	3,060	667	0.22
14	A81016	9,630	2,160	0.22
15	A81017	20,746	4,339	0.21
16	A81018	5,473	1,633	0.30
17	A81019	10,381	2,253	0.22
18	A81020	7,907	2,558	0.32
19	A81021	13,189	4,043	0.31
20	A81022	10,743	2,539	0.24
21	A81023	7,674	1,182	0.15
22	A81025	4,162	1,863	0.45
23	A81026	13,758	2,661	0.19
24	A81027	13,255	2,836	0.21
25	A81029	5,799	1,370	0.24
26	A81030	12,119	3,509	0.29
27	A81031	12,729	2,984	0.23
28	A81032	10,505	2,153	0.20
29	A81033	3,371	822	0.24
30	A81034	21,774	5,148	0.24
31	A81035	10,751	2,067	0.19
32	A81036	17,362	4,589	0.26
33	A81037	7,045	2,239	0.32
34	A81038	3,311	716	0.22
35	A81039	9,359	2,411	0.26
36	A81040	8,552	2,212	0.26
37	A81041	9,088	2,813	0.31
38	A81042	6,379	1,846	0.29
39	A81043	8,002	1,611	0.20

	REGION	MORTALITY RATES	REGIONAL STATIN SPEND PER PATIENT
0	East Midlands	0.0739%	£0.22
1	East of England	0.0632%	£0.21
2	London	0.0492%	£0.17
3	North East	0.0834%	£0.26
4	North West	0.0821%	£0.25
5	South East	0.0571%	£0.22
6	South West	0.0672%	£0.24
7	West Midlands	0.0717%	£0.23
8	Yorkshire and The Humber	0.0762%	£0.26

	Min. annual cases of flu (Uruguay min.)	Min. annual cases of flu (Romania min.)	Max. annual cases of flu (Uruguay max.)	Max. annual cases of flu (Romania max.)
year
2003	102	490	308	1,434
2004	98	582	251	932
2005	81	591	205	1,277
2006	75	552	197	1,293
2007	71	507	294	1,497
2008	70	529	210	1,305
2009	54	535	282	1,151
2010	51	445	172	880
2011	52	464	175	1,143
2012	49	471	200	950
2013	55	502	159	971
2014	56	515	204	966
2015	70	502	225	1,248